fix(mcp): disable MCP_SKILLS feature flag — source not mirrored

Closes #856.

MCP servers that expose resources (e.g. RepoPrompt) failed to load
their tools in the open build with:

    Error fetching tools/commands/resources:
    fetchMcpSkillsForClient is not a function

Root cause: scripts/build.ts set MCP_SKILLS: true, which made
feature('MCP_SKILLS') evaluate to true at build time. The guards
around the dynamic skill discovery path therefore stayed live. The
underlying source file src/skills/mcpSkills.ts is not mirrored into
the open tree, so the bundler fell back to its generic missing-module
stub — which only exports `default` for require()-style imports, not
the named `fetchMcpSkillsForClient` binding. At runtime the require
returned an object without that property, and calling it threw.

`openclaude mcp doctor` reported RepoPrompt as healthy because doctor
does not exercise the skills-fetch path.

Fix: flip MCP_SKILLS to false and move it into the "Disabled: missing
source" group. With the flag off, every `if (feature('MCP_SKILLS'))`
guard becomes a no-op at build time, the require() branch is dead
code, and MCP servers with resources load normally via the existing
`Promise.resolve([])` fallbacks already present at each call site.

Also adds scripts/feature-flags-source-guard.test.ts to fail fast if
MCP_SKILLS (or any future flag in the same category) is re-enabled
without the corresponding source file being mirrored first.

Verification:
  - Test fails on main, passes with this fix
  - `bun run build` produces a bundle with no
    `missing-module-stub:../../skills/mcpSkills.js` reference
  - Full `bun test` — 1222 pass / 12 fail (same pre-existing 12 as
    main; new test adds the +1 pass)

.env.example

@@ -145,27 +145,9 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # CLAUDE_CODE_USE_OPENAI=1
 # OPENAI_API_KEY=sk-your-key-here
 # OPENAI_MODEL=gpt-4o
-# For DeepSeek, set:
-# OPENAI_BASE_URL=https://api.deepseek.com/v1
-# OPENAI_MODEL=deepseek-v4-flash
-# Optional: OPENAI_MODEL=deepseek-v4-pro
-# Legacy aliases also work: deepseek-chat and deepseek-reasoner
-# For Z.AI GLM Coding Plan, set:
-# OPENAI_BASE_URL=https://api.z.ai/api/coding/paas/v4
-# OPENAI_MODEL=GLM-5.1
-# Optional: OPENAI_MODEL=GLM-5-Turbo, GLM-4.7, or GLM-4.5-Air
 
 # Use a custom OpenAI-compatible endpoint (optional — defaults to api.openai.com)
 # OPENAI_BASE_URL=https://api.openai.com/v1
-# Choose the OpenAI-compatible API surface (optional — defaults to chat_completions)
-# Supported: chat_completions, responses
-# OPENAI_API_FORMAT=chat_completions
-# Choose a custom auth header for OpenAI-compatible providers (optional).
-# Authorization defaults to Bearer; custom headers default to the raw API key.
-# Set OPENAI_AUTH_HEADER_VALUE when the header value differs from OPENAI_API_KEY.
-# OPENAI_AUTH_HEADER=api-key
-# OPENAI_AUTH_SCHEME=raw
-# OPENAI_AUTH_HEADER_VALUE=your-header-value-here
 
 # Fallback context window size (tokens) when the model is not found in the
 # built-in table (default: 128000). Increase this for models with larger
@@ -312,20 +294,6 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # Useful for users who want full transparency over what the model sees
 # OPENCLAUDE_DISABLE_TOOL_REMINDERS=1
 
-# Log structured per-request token usage (including cache metrics) to stderr.
-# Useful for auditing cache hit rate / debugging cost spikes outside the REPL.
-# Any truthy value enables it ("verbose", "1", "true").
-#
-# Complements (does NOT replace) CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT —
-# they serve different audiences:
-#   - OPENCLAUDE_LOG_TOKEN_USAGE is user-facing: one JSON line per API
-#     request on stderr, intended for humans inspecting cost/caching.
-#   - CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT is model-facing: injects
-#     a context-usage attachment INTO the prompt so the model can reason
-#     about its own remaining context. Does not touch stderr.
-# Turn on whichever audience you're debugging; both can run together.
-# OPENCLAUDE_LOG_TOKEN_USAGE=verbose
-
 # Custom timeout for API requests in milliseconds (default: varies)
 # API_TIMEOUT_MS=60000
 

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "0.7.0"
+  ".": "0.6.0"
 }

CHANGELOG.md

@@ -1,39 +1,5 @@
 # Changelog
 
-## [0.7.0](https://github.com/Gitlawb/openclaude/compare/v0.6.0...v0.7.0) (2026-04-26)
-
-
-### Features
-
-* add model-specific tokenizers and compression ratio detection ([#799](https://github.com/Gitlawb/openclaude/issues/799)) ([e92e527](https://github.com/Gitlawb/openclaude/commit/e92e5274b223d935d380b1fbd234cb631ab03211))
-* add OPENCLAUDE_DISABLE_TOOL_REMINDERS env var to suppress hidden tool-output reminders ([#837](https://github.com/Gitlawb/openclaude/issues/837)) ([28de94d](https://github.com/Gitlawb/openclaude/commit/28de94df5dcd7718cb334e2e793e9472f5b291c5)), closes [#809](https://github.com/Gitlawb/openclaude/issues/809)
-* add streaming optimizer and structured request logging ([#703](https://github.com/Gitlawb/openclaude/issues/703)) ([5b9cd21](https://github.com/Gitlawb/openclaude/commit/5b9cd21e373823a77fd552d6e02f5d4b68ae06b1))
-* add xAI as official provider ([#865](https://github.com/Gitlawb/openclaude/issues/865)) ([2586a9c](https://github.com/Gitlawb/openclaude/commit/2586a9cddbd2512826bca81cb5deb3ec97f00f0f))
-* **api:** expose cache metrics in REPL + normalize across providers ([#813](https://github.com/Gitlawb/openclaude/issues/813)) ([9e23c2b](https://github.com/Gitlawb/openclaude/commit/9e23c2bec43697187762601db5b1585c9b0fb1a3))
-* implement Hook Chains runtime integration for self-healing agent mesh MVP ([#711](https://github.com/Gitlawb/openclaude/issues/711)) ([44a2c30](https://github.com/Gitlawb/openclaude/commit/44a2c30d5f9b98027e454466c680360f6b4625fc))
-* **memory:** implement persistent project-level Knowledge Graph and RAG ([#899](https://github.com/Gitlawb/openclaude/issues/899)) ([29f7579](https://github.com/Gitlawb/openclaude/commit/29f757937732be0f8cca2bc0627a27eeafc2a992))
-* **minimax:** add /usage support and fix MiniMax quota parsing ([#869](https://github.com/Gitlawb/openclaude/issues/869)) ([26413f6](https://github.com/Gitlawb/openclaude/commit/26413f6d307928a4f14c9c61c9860a28f8d81358))
-* **model:** add GPT-5.5 support for Codex provider ([#880](https://github.com/Gitlawb/openclaude/issues/880)) ([038f715](https://github.com/Gitlawb/openclaude/commit/038f715b7ab9714340bda421b73a86d8590cf531))
-* **tools:** resilient web search and fetch across all providers ([#836](https://github.com/Gitlawb/openclaude/issues/836)) ([531e3f1](https://github.com/Gitlawb/openclaude/commit/531e3f10592a73d81f26675c2479d46a3d5b55f5))
-* **zai:** add Z.AI GLM Coding Plan provider preset ([#896](https://github.com/Gitlawb/openclaude/issues/896)) ([a0d657e](https://github.com/Gitlawb/openclaude/commit/a0d657ee188f52f8a4ceaad1658c81343a32fdad))
-
-
-### Bug Fixes
-
-* **agent:** provider-aware fallback for haiku/sonnet aliases ([#908](https://github.com/Gitlawb/openclaude/issues/908)) ([a3e728a](https://github.com/Gitlawb/openclaude/commit/a3e728a114f6379b80daefc8abcac17a752c5f96))
-* bugs ([#885](https://github.com/Gitlawb/openclaude/issues/885)) ([c6c5f06](https://github.com/Gitlawb/openclaude/commit/c6c5f0608cf6509b412b121954547d72b3f3a411))
-* make OpenAI fallback context window configurable + support external model lookup ([#861](https://github.com/Gitlawb/openclaude/issues/861)) ([b750e9e](https://github.com/Gitlawb/openclaude/commit/b750e9e97d15926d094d435772b2d6d12e5e545c))
-* **mcp:** disable MCP_SKILLS feature flag — source not mirrored ([#872](https://github.com/Gitlawb/openclaude/issues/872)) ([dcbe295](https://github.com/Gitlawb/openclaude/commit/dcbe29558ab9c74d335b138488005a6509aa906a))
-* normalize /provider multi-model selection and semicolon parsing ([#841](https://github.com/Gitlawb/openclaude/issues/841)) ([c4cb98a](https://github.com/Gitlawb/openclaude/commit/c4cb98a4f092062da02a4728cf59fed0fc3a6d3f))
-* **openai-shim:** echo reasoning_content on assistant tool-call messages for Moonshot ([#828](https://github.com/Gitlawb/openclaude/issues/828)) ([67de6bd](https://github.com/Gitlawb/openclaude/commit/67de6bd2cffc3381f0f28fd3ffce043970611667))
-* **query:** restore system prompt structure and add missing config import ([#907](https://github.com/Gitlawb/openclaude/issues/907)) ([818689b](https://github.com/Gitlawb/openclaude/commit/818689b2ee71cb6966cb4dc5a5ebd90fd22b0fcb))
-* **shell:** recover when CWD path was replaced by a non-directory ([#871](https://github.com/Gitlawb/openclaude/issues/871)) ([a4c6757](https://github.com/Gitlawb/openclaude/commit/a4c67570238794317d049a225396672b465fdbfc))
-* **startup:** show --model flag override on startup screen ([#898](https://github.com/Gitlawb/openclaude/issues/898)) ([d45628c](https://github.com/Gitlawb/openclaude/commit/d45628c41300b83b466e6a97983099615a50e7d7))
-* **startup:** url authoritative over model name in banner provider detect ([#864](https://github.com/Gitlawb/openclaude/issues/864)) ([e346b8d](https://github.com/Gitlawb/openclaude/commit/e346b8d5ec2d58a4e8db337918d52d844ee52766)), closes [#855](https://github.com/Gitlawb/openclaude/issues/855)
-* surface actionable error when DuckDuckGo web search is rate-limited ([#834](https://github.com/Gitlawb/openclaude/issues/834)) ([3c4d843](https://github.com/Gitlawb/openclaude/commit/3c4d8435c42e1ee04f9defd31c4c589017f524c5))
-* **test:** add missing teammate exports to hookChains integration mock ([#840](https://github.com/Gitlawb/openclaude/issues/840)) ([23e8cfb](https://github.com/Gitlawb/openclaude/commit/23e8cfbd5b22179684276bef4131e26b830ce69c)), closes [#839](https://github.com/Gitlawb/openclaude/issues/839)
-* **update:** show real package version and give actionable guidance ([#870](https://github.com/Gitlawb/openclaude/issues/870)) ([6e58b81](https://github.com/Gitlawb/openclaude/commit/6e58b819370128b923dda4fcc774bb556f4b951a))
-
 ## [0.6.0](https://github.com/Gitlawb/openclaude/compare/v0.5.2...v0.6.0) (2026-04-22)
 
 

PLAYBOOK.md

@@ -132,7 +132,7 @@ Cause:
 Fix:
 
 ```powershell
-cd <PATH>
+cd C:\Users\Lucas Pedry\Documents\openclaude\openclaude
 bun run dev:profile
 ```
 
@@ -189,7 +189,7 @@ Or pick a local Ollama profile automatically by goal:
 bun run profile:init -- --provider ollama --goal balanced
 ```
 
-## 6.5 Placeholder key (`YOUR_KEY`) error
+## 6.5 Placeholder key (`SUA_CHAVE`) error
 
 Cause:
 

README.md

@@ -13,25 +13,7 @@ Use OpenAI-compatible APIs, Gemini, GitHub Models, Codex OAuth, Codex, Ollama, A
 OpenClaude is also mirrored to GitLawb:
 [gitlawb.com/node/repos/z6MkqDnb/openclaude](https://gitlawb.com/node/repos/z6MkqDnb/openclaude)
 
-[Quick Start](#quick-start) | [Setup Guides](#setup-guides) | [Providers](#supported-providers) | [Source Build](#source-build-and-local-development) | [VS Code Extension](#vs-code-extension) | [Sponsors](#sponsors) | [Community](#community)
-
-## Sponsors
-
-<p align="center">
-  <a href="https://gitlawb.com">
-    <img src="https://gitlawb.com/logo.png" alt="GitLawb logo" width="96">
-  </a>
-  &nbsp;&nbsp;&nbsp;&nbsp;
-  <a href="https://bankr.bot">
-    <img src="https://bankr.bot/favicon.svg" alt="Bankr.bot logo" width="96">
-  </a>
-</p>
-
-<p align="center">
-  <a href="https://gitlawb.com"><strong>GitLawb</strong></a>
-  &nbsp;&nbsp;&nbsp;&nbsp;
-  <a href="https://bankr.bot"><strong>Bankr.bot</strong></a>
-</p>
+[Quick Start](#quick-start) | [Setup Guides](#setup-guides) | [Providers](#supported-providers) | [Source Build](#source-build-and-local-development) | [VS Code Extension](#vs-code-extension) | [Community](#community)
 
 ## Star History
 
@@ -170,12 +152,12 @@ For best results, use models with strong tool/function calling support.
 
 OpenClaude can route different agents to different models through settings-based routing. This is useful for cost optimization or splitting work by model strength.
 
-Add to `~/.openclaude.json`:
+Add to `~/.claude/settings.json`:
 
 ```json
 {
   "agentModels": {
-    "deepseek-v4-flash": {
+    "deepseek-chat": {
       "base_url": "https://api.deepseek.com/v1",
       "api_key": "sk-your-key"
     },
@@ -185,10 +167,10 @@ Add to `~/.openclaude.json`:
     }
   },
   "agentRouting": {
-    "Explore": "deepseek-v4-flash",
+    "Explore": "deepseek-chat",
     "Plan": "gpt-4o",
     "general-purpose": "gpt-4o",
-    "frontend-dev": "deepseek-v4-flash",
+    "frontend-dev": "deepseek-chat",
     "default": "gpt-4o"
   }
 }

docs/advanced-setup.md

@@ -68,11 +68,9 @@ openclaude
 export CLAUDE_CODE_USE_OPENAI=1
 export OPENAI_API_KEY=sk-...
 export OPENAI_BASE_URL=https://api.deepseek.com/v1
-export OPENAI_MODEL=deepseek-v4-flash
+export OPENAI_MODEL=deepseek-chat
 ```
 
-Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` remain available as DeepSeek's legacy API aliases.
-
 ### Google Gemini via OpenRouter
 
 ```bash
@@ -171,13 +169,12 @@ export OPENAI_MODEL=gpt-4o
 |----------|----------|-------------|
 | `CLAUDE_CODE_USE_OPENAI` | Yes | Set to `1` to enable the OpenAI provider |
 | `OPENAI_API_KEY` | Yes* | Your API key (`*` not needed for local models like Ollama or Atomic Chat) |
-| `OPENAI_MODEL` | Yes | Model name such as `gpt-4o`, `deepseek-v4-flash`, or `llama3.3:70b` |
+| `OPENAI_MODEL` | Yes | Model name such as `gpt-4o`, `deepseek-chat`, or `llama3.3:70b` |
 | `OPENAI_BASE_URL` | No | API endpoint, defaulting to `https://api.openai.com/v1` |
 | `CODEX_API_KEY` | Codex only | Codex or ChatGPT access token override |
 | `CODEX_AUTH_JSON_PATH` | Codex only | Path to a Codex CLI `auth.json` file |
 | `CODEX_HOME` | Codex only | Alternative Codex home directory |
 | `OPENCLAUDE_DISABLE_CO_AUTHORED_BY` | No | Suppress the default `Co-Authored-By` trailer in generated git commits |
-| `OPENCLAUDE_LOG_TOKEN_USAGE` | No | When truthy (e.g. `verbose`), emits one JSON line on stderr per API request with input/output/cache tokens and the resolved provider. **User-facing debug output** — complements the REPL display controlled by `/config showCacheStats`. Distinct from `CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT`, which is **model-facing** (injects context usage info into the prompt itself). Both can run together. |
 
 You can also use `ANTHROPIC_MODEL` to override the model name. `OPENAI_MODEL` takes priority.
 

docs/quick-start-mac-linux.md

@@ -41,13 +41,11 @@ openclaude
 export CLAUDE_CODE_USE_OPENAI=1
 export OPENAI_API_KEY=sk-your-key-here
 export OPENAI_BASE_URL=https://api.deepseek.com/v1
-export OPENAI_MODEL=deepseek-v4-flash
+export OPENAI_MODEL=deepseek-chat
 
 openclaude
 ```
 
-Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` still work as DeepSeek's legacy API aliases.
-
 ### Option C: Ollama
 
 Install Ollama first from:

docs/quick-start-windows.md

@@ -41,13 +41,11 @@ openclaude
 $env:CLAUDE_CODE_USE_OPENAI="1"
 $env:OPENAI_API_KEY="sk-your-key-here"
 $env:OPENAI_BASE_URL="https://api.deepseek.com/v1"
-$env:OPENAI_MODEL="deepseek-v4-flash"
+$env:OPENAI_MODEL="deepseek-chat"
 
 openclaude
 ```
 
-Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` still work as DeepSeek's legacy API aliases.
-
 ### Option C: Ollama
 
 Install Ollama first from:

package.json

@@ -1,7 +1,7 @@
 {
   "name": "@gitlawb/openclaude",
-  "version": "0.7.0",
-  "description": "OpenClaude opens coding-agent workflows to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models",
+  "version": "0.6.0",
+  "description": "Claude Code opened to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models",
   "type": "module",
   "bin": {
     "openclaude": "./bin/openclaude"

src/bridge/bridgeEnabled.ts

@@ -70,13 +70,13 @@ export async function isBridgeEnabledBlocking(): Promise<boolean> {
 export async function getBridgeDisabledReason(): Promise<string | null> {
   if (feature('BRIDGE_MODE')) {
     if (!isClaudeAISubscriber()) {
-      return 'Remote Control requires a claude.ai subscription. Run `openclaude auth login` to sign in with your claude.ai account.'
+      return 'Remote Control requires a claude.ai subscription. Run `claude auth login` to sign in with your claude.ai account.'
     }
     if (!hasProfileScope()) {
-      return 'Remote Control requires a full-scope login token. Long-lived tokens (from `openclaude setup-token` or CLAUDE_CODE_OAUTH_TOKEN) are limited to inference-only for security reasons. Run `openclaude auth login` to use Remote Control.'
+      return 'Remote Control requires a full-scope login token. Long-lived tokens (from `claude setup-token` or CLAUDE_CODE_OAUTH_TOKEN) are limited to inference-only for security reasons. Run `claude auth login` to use Remote Control.'
     }
     if (!getOauthAccountInfo()?.organizationUuid) {
-      return 'Unable to determine your organization for Remote Control eligibility. Run `openclaude auth login` to refresh your account information.'
+      return 'Unable to determine your organization for Remote Control eligibility. Run `claude auth login` to refresh your account information.'
     }
     if (!(await checkGate_CACHED_OR_BLOCKING('tengu_ccr_bridge'))) {
       return 'Remote Control is not yet enabled for your account.'
@@ -166,7 +166,7 @@ export function checkBridgeMinVersion(): string | null {
       minVersion: string
     }>('tengu_bridge_min_version', { minVersion: '0.0.0' })
     if (config.minVersion && lt(MACRO.VERSION, config.minVersion)) {
-      return `Your version of OpenClaude (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${config.minVersion} or higher is required. Run \`openclaude update\` to update.`
+      return `Your version of Claude Code (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${config.minVersion} or higher is required. Run \`claude update\` to update.`
     }
   }
   return null

src/bridge/bridgeMain.ts

@@ -2248,7 +2248,7 @@ export async function bridgeMain(args: string[]): Promise<void> {
     })
     // biome-ignore lint/suspicious/noConsole: intentional dialog output
     console.log(
-      `\nClaude Remote Control is launching in spawn mode which lets you create new sessions in this project from OpenClaude on the web or your mobile app. Learn more here: https://code.claude.com/docs/en/remote-control\n\n` +
+      `\nClaude Remote Control is launching in spawn mode which lets you create new sessions in this project from Claude Code on Web or your Mobile app. Learn more here: https://code.claude.com/docs/en/remote-control\n\n` +
         `Spawn mode for this project:\n` +
         `  [1] same-dir \u2014 sessions share the current directory (default)\n` +
         `  [2] worktree \u2014 each session gets an isolated git worktree\n\n` +

src/bridge/envLessBridgeConfig.ts

@@ -147,7 +147,7 @@ export async function getEnvLessBridgeConfig(): Promise<EnvLessBridgeConfig> {
 export async function checkEnvLessBridgeMinVersion(): Promise<string | null> {
   const cfg = await getEnvLessBridgeConfig()
   if (cfg.min_version && lt(MACRO.VERSION, cfg.min_version)) {
-    return `Your version of OpenClaude (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${cfg.min_version} or higher is required. Run \`openclaude update\` to update.`
+    return `Your version of Claude Code (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${cfg.min_version} or higher is required. Run \`claude update\` to update.`
   }
   return null
 }

src/bridge/initReplBridge.ts

@@ -415,7 +415,7 @@ export async function initReplBridge(
         `[bridge:repl] Skipping: ${versionError}`,
         true,
       )
-      onStateChange?.('failed', 'run `openclaude update` to upgrade')
+      onStateChange?.('failed', 'run `claude update` to upgrade')
       return null
     }
     logForDebugging(
@@ -456,7 +456,7 @@ export async function initReplBridge(
   const versionError = checkBridgeMinVersion()
   if (versionError) {
     logBridgeSkip('version_too_old', `[bridge:repl] Skipping: ${versionError}`)
-    onStateChange?.('failed', 'run `openclaude update` to upgrade')
+    onStateChange?.('failed', 'run `claude update` to upgrade')
     return null
   }
 

src/bridge/trustedDevice.ts

@@ -147,7 +147,7 @@ export async function enrollTrustedDevice(): Promise<void> {
         device_id?: string
       }>(
         `${baseUrl}/api/auth/trusted_devices`,
-        { display_name: `OpenClaude on ${hostname()} · ${process.platform}` },
+        { display_name: `Claude Code on ${hostname()} · ${process.platform}` },
         {
           headers: {
             Authorization: `Bearer ${accessToken}`,

src/cli/handlers/auth.ts

@@ -287,7 +287,7 @@ export async function authStatus(opts: {
     }
     if (!loggedIn) {
       process.stdout.write(
-        'Not logged in. Run openclaude auth login to authenticate.\n',
+        'Not logged in. Run claude auth login to authenticate.\n',
       )
     }
   } else {

src/cli/handlers/autoMode.ts

@@ -83,7 +83,7 @@ export async function autoModeCritiqueHandler(options: {
     process.stdout.write(
       'No custom auto mode rules found.\n\n' +
         'Add rules to your settings file under autoMode.{allow, soft_deny, environment}.\n' +
-        'Run `openclaude auto-mode defaults` to see the default rules for reference.\n',
+        'Run `claude auto-mode defaults` to see the default rules for reference.\n',
     )
     return
   }

src/cli/handlers/mcp.tsx

@@ -233,7 +233,7 @@ export async function mcpRemoveHandler(name: string, options: {
       });
       process.stderr.write('\nTo remove from a specific scope, use:\n');
       scopes.forEach(scope => {
-        process.stderr.write(`  openclaude mcp remove "${name}" -s ${scope}\n`);
+        process.stderr.write(`  claude mcp remove "${name}" -s ${scope}\n`);
       });
       cliError();
     }
@@ -250,7 +250,7 @@ export async function mcpListHandler(): Promise<void> {
   } = await getAllMcpConfigs();
   if (Object.keys(configs).length === 0) {
     // biome-ignore lint/suspicious/noConsole:: intentional console output
-    console.log('No MCP servers configured. Use `openclaude mcp add` to add a server.');
+    console.log('No MCP servers configured. Use `claude mcp add` to add a server.');
   } else {
     // biome-ignore lint/suspicious/noConsole:: intentional console output
     console.log('Checking MCP server health...\n');
@@ -374,7 +374,7 @@ export async function mcpGetHandler(name: string): Promise<void> {
     }
   }
   // biome-ignore lint/suspicious/noConsole:: intentional console output
-  console.log(`\nTo remove this server, run: openclaude mcp remove "${name}" -s ${server.scope}`);
+  console.log(`\nTo remove this server, run: claude mcp remove "${name}" -s ${server.scope}`);
   // Use gracefulShutdown to properly clean up MCP server connections
   // (process.exit bypasses cleanup handlers, leaving child processes orphaned)
   await gracefulShutdown(0);
@@ -455,5 +455,5 @@ export async function mcpResetChoicesHandler(): Promise<void> {
     disabledMcpjsonServers: [],
     enableAllProjectMcpServers: false
   }));
-  cliOk('All project-scoped (.mcp.json) server approvals and rejections have been reset.\n' + 'You will be prompted for approval next time you start OpenClaude.');
+  cliOk('All project-scoped (.mcp.json) server approvals and rejections have been reset.\n' + 'You will be prompted for approval next time you start Claude Code.');
 }

src/cli/handlers/plugins.ts

@@ -352,7 +352,7 @@ export async function pluginListHandler(options: {
     // through to the session section so the failure is visible.
     if (inlineLoadErrors.length === 0) {
       cliOk(
-        'No plugins installed. Use `openclaude plugin install` to install a plugin.',
+        'No plugins installed. Use `claude plugin install` to install a plugin.',
       )
     }
   }

src/cli/print.ts

@@ -5026,7 +5026,7 @@ async function loadInitialMessages(
       )
       if (!parsedSessionId) {
         let errorMessage =
-          'Error: --resume requires a valid session ID when used with --print. Usage: openclaude -p --resume <session-id>'
+          'Error: --resume requires a valid session ID when used with --print. Usage: claude -p --resume <session-id>'
         if (typeof options.resume === 'string') {
           errorMessage += `. Session IDs must be in UUID format (e.g., 550e8400-e29b-41d4-a716-446655440000). Provided value "${options.resume}" is not a valid UUID`
         }

src/cli/update.ts

@@ -35,20 +35,15 @@ export async function update() {
   // binary (without it).
   if (getAPIProvider() !== 'firstParty') {
     writeToStdout(
-      chalk.yellow(
-        `Auto-update is not available for third-party provider builds.\n`,
-      ) +
-        `Current version: ${MACRO.DISPLAY_VERSION}\n\n` +
-        `To update, reinstall from npm:\n` +
-        chalk.bold(`  npm install -g ${MACRO.PACKAGE_URL}@latest`) + '\n\n' +
-        `Or, if you built from source, pull and rebuild:\n` +
-        chalk.bold('  git pull && bun install && bun run build') + '\n',
+      chalk.yellow('Auto-update is not available for third-party provider builds.\n') +
+      'To update, pull the latest source from the repository and rebuild:\n' +
+      '  git pull && bun install && bun run build\n',
     )
-    await gracefulShutdown(0)
+    return
   }
 
   logEvent('tengu_update_check', {})
-  writeToStdout(`Current version: ${MACRO.DISPLAY_VERSION}\n`)
+  writeToStdout(`Current version: ${MACRO.VERSION}\n`)
 
   const channel = getInitialSettings()?.autoUpdatesChannel ?? 'latest'
   writeToStdout(`Checking for updates to ${channel} version...\n`)
@@ -128,14 +123,9 @@ export async function update() {
   if (diagnostic.installationType === 'development') {
     writeToStdout('\n')
     writeToStdout(
-      chalk.yellow('You are running a development build — auto-update is unavailable.') + '\n',
+      chalk.yellow('Warning: Cannot update development build') + '\n',
     )
-    writeToStdout('To update, pull the latest source and rebuild:\n')
-    writeToStdout(chalk.bold('  git pull && bun install && bun run build') + '\n')
-    writeToStdout('\n')
-    writeToStdout('Or reinstall from npm:\n')
-    writeToStdout(chalk.bold(`  npm install -g ${MACRO.PACKAGE_URL}@latest`) + '\n')
-    await gracefulShutdown(0)
+    await gracefulShutdown(1)
   }
 
   // Check if running from a package manager
@@ -146,8 +136,8 @@ export async function update() {
     if (packageManager === 'homebrew') {
       writeToStdout('Claude is managed by Homebrew.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(chalk.bold('  brew upgrade claude-code') + '\n')
@@ -157,8 +147,8 @@ export async function update() {
     } else if (packageManager === 'winget') {
       writeToStdout('Claude is managed by winget.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(
@@ -170,8 +160,8 @@ export async function update() {
     } else if (packageManager === 'apk') {
       writeToStdout('Claude is managed by apk.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(chalk.bold('  apk upgrade claude-code') + '\n')
@@ -260,14 +250,14 @@ export async function update() {
         await gracefulShutdown(1)
       }
 
-      if (result.latestVersion === MACRO.DISPLAY_VERSION) {
+      if (result.latestVersion === MACRO.VERSION) {
         writeToStdout(
-          chalk.green(`OpenClaude is up to date (${MACRO.DISPLAY_VERSION})`) + '\n',
+          chalk.green(`Claude Code is up to date (${MACRO.VERSION})`) + '\n',
         )
       } else {
         writeToStdout(
           chalk.green(
-            `Successfully updated from ${MACRO.DISPLAY_VERSION} to version ${result.latestVersion}`,
+            `Successfully updated from ${MACRO.VERSION} to version ${result.latestVersion}`,
           ) + '\n',
         )
         await regenerateCompletionCache()
@@ -276,7 +266,7 @@ export async function update() {
     } catch (error) {
       process.stderr.write('Error: Failed to install native update\n')
       process.stderr.write(String(error) + '\n')
-      process.stderr.write('Try running "openclaude doctor" for diagnostics\n')
+      process.stderr.write('Try running "claude doctor" for diagnostics\n')
       await gracefulShutdown(1)
     }
   }
@@ -330,15 +320,15 @@ export async function update() {
   }
 
   // Check if versions match exactly, including any build metadata (like SHA)
-  if (latestVersion === MACRO.DISPLAY_VERSION) {
+  if (latestVersion === MACRO.VERSION) {
     writeToStdout(
-      chalk.green(`OpenClaude is up to date (${MACRO.DISPLAY_VERSION})`) + '\n',
+      chalk.green(`Claude Code is up to date (${MACRO.VERSION})`) + '\n',
     )
     await gracefulShutdown(0)
   }
 
   writeToStdout(
-    `New version available: ${latestVersion} (current: ${MACRO.DISPLAY_VERSION})\n`,
+    `New version available: ${latestVersion} (current: ${MACRO.VERSION})\n`,
   )
   writeToStdout('Installing update...\n')
 
@@ -398,7 +388,7 @@ export async function update() {
     case 'success':
       writeToStdout(
         chalk.green(
-          `Successfully updated from ${MACRO.DISPLAY_VERSION} to version ${latestVersion}`,
+          `Successfully updated from ${MACRO.VERSION} to version ${latestVersion}`,
         ) + '\n',
       )
       await regenerateCompletionCache()

src/commands.ts

@@ -21,7 +21,6 @@ import dream from './commands/dream/index.js'
 import ctx_viz from './commands/ctx_viz/index.js'
 import doctor from './commands/doctor/index.js'
 import onboardGithub from './commands/onboard-github/index.js'
-import knowledge from './commands/knowledge/index.js'
 import memory from './commands/memory/index.js'
 import help from './commands/help/index.js'
 import ide from './commands/ide/index.js'
@@ -34,7 +33,6 @@ import installGitHubApp from './commands/install-github-app/index.js'
 import installSlackApp from './commands/install-slack-app/index.js'
 import breakCache from './commands/break-cache/index.js'
 import cacheProbe from './commands/cache-probe/index.js'
-import cacheStats from './commands/cacheStats/index.js'
 import mcp from './commands/mcp/index.js'
 import mobile from './commands/mobile/index.js'
 import onboarding from './commands/onboarding/index.js'
@@ -199,7 +197,7 @@ import stats from './commands/stats/index.js'
 const usageReport: Command = {
   type: 'prompt',
   name: 'insights',
-  description: 'Generate a report analyzing your OpenClaude sessions',
+  description: 'Generate a report analyzing your Claude Code sessions',
   contentLength: 0,
   progressMessage: 'analyzing your sessions',
   source: 'builtin',
@@ -272,7 +270,6 @@ const COMMANDS = memoize((): Command[] => [
   branch,
   btw,
   cacheProbe,
-  cacheStats,
   chrome,
   clear,
   color,
@@ -295,7 +292,6 @@ const COMMANDS = memoize((): Command[] => [
   ide,
   init,
   keybindings,
-  knowledge,
   installGitHubApp,
   installSlackApp,
   mcp,

src/commands/buddy/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const buddy = {
   type: 'local-jsx',
   name: 'buddy',
-  description: 'Hatch, pet, and manage your OpenClaude companion',
+  description: 'Hatch, pet, and manage your Open Claude companion',
   immediate: true,
   argumentHint: '[status|mute|unmute|help]',
   load: () => import('./buddy.js'),

src/commands/cacheStats/cacheStats.test.ts

@@ -1,157 +0,0 @@
-/**
- * Tests for `/cache-stats` command rendering.
- *
- * The command has non-trivial string formatting (timestamp slicing, model
- * label padding, conditional N/A footnote, recent-rows cap) which can
- * silently regress — these snapshot tests keep it honest.
- */
-import { beforeEach, describe, expect, test } from 'bun:test'
-import type { CacheMetrics } from '../../services/api/cacheMetrics.js'
-import {
-  _setHistoryCapForTesting,
-  recordRequest,
-  resetSessionCacheStats,
-} from '../../services/api/cacheStatsTracker.js'
-import { call } from './cacheStats.js'
-
-function supported(partial: Partial<CacheMetrics>): CacheMetrics {
-  return {
-    read: 0,
-    created: 0,
-    total: 0,
-    hitRate: null,
-    supported: true,
-    ...partial,
-  }
-}
-
-const UNSUPPORTED: CacheMetrics = {
-  read: 0,
-  created: 0,
-  total: 0,
-  hitRate: null,
-  supported: false,
-}
-
-// The command signature requires a LocalJSXCommandContext. Our command
-// doesn't actually read it — we pass an empty stand-in so the test can
-// invoke call() without dragging the whole REPL context in.
-const EMPTY_CTX = {} as Parameters<typeof call>[1]
-
-// /cache-stats always returns a text result. Narrow the union here so
-// the assertions don't need to redo the discriminant check every call.
-async function runCommand(): Promise<string> {
-  const result = await call('', EMPTY_CTX)
-  if (result.type !== 'text') {
-    throw new Error(
-      `cacheStats command must return type:'text', got ${result.type}`,
-    )
-  }
-  return result.value
-}
-
-beforeEach(() => {
-  resetSessionCacheStats()
-  _setHistoryCapForTesting(500)
-})
-
-describe('/cache-stats — empty session', () => {
-  test('shows friendly "no requests yet" message', async () => {
-    const value = await runCommand()
-    expect(value).toContain('No API requests yet this session')
-    expect(value).toContain('/cache-stats')
-  })
-})
-
-describe('/cache-stats — supported-only session', () => {
-  test('renders Cache stats header, turn and session summaries', async () => {
-    recordRequest(
-      supported({ read: 500, total: 1_000, hitRate: 0.5 }),
-      'claude-sonnet-4',
-    )
-    const value = await runCommand()
-    expect(value).toContain('Cache stats')
-    expect(value).toContain('Current turn:')
-    expect(value).toContain('Session total:')
-    // Compact metric line should appear in the recent-requests table.
-    expect(value).toContain('claude-sonnet-4')
-    expect(value).toContain('read')
-  })
-
-  test('omits the N/A footnote when every row is supported', async () => {
-    recordRequest(supported({ read: 200, total: 400, hitRate: 0.5 }), 'model-A')
-    const value = await runCommand()
-    expect(value).not.toContain('N/A rows')
-  })
-})
-
-describe('/cache-stats — mixed supported + unsupported', () => {
-  test('renders N/A footnote when any row is unsupported', async () => {
-    recordRequest(UNSUPPORTED, 'gpt-4-copilot')
-    recordRequest(
-      supported({ read: 100, total: 500, hitRate: 0.2 }),
-      'claude-sonnet-4',
-    )
-    const value = await runCommand()
-    expect(value).toContain(
-      'N/A rows: provider API does not expose cache usage',
-    )
-    expect(value).toContain('GitHub Copilot')
-    expect(value).toContain('Ollama')
-  })
-})
-
-describe('/cache-stats — recent-rows cap', () => {
-  test('caps the breakdown at 20 rows and reports omitted count', async () => {
-    for (let i = 0; i < 25; i++) {
-      recordRequest(
-        supported({ read: i, total: 100, hitRate: i / 100 }),
-        `model-${i}`,
-      )
-    }
-    const value = await runCommand()
-    // 20 shown, 5 omitted from the oldest end.
-    expect(value).toContain('(20 of 25, 5 older omitted)')
-    // Oldest rows (model-0..model-4) should not appear; newest must.
-    expect(value).toContain('model-24')
-    expect(value).not.toContain('model-0 ')
-  })
-
-  test('does not mention "older omitted" when all rows fit', async () => {
-    for (let i = 0; i < 5; i++) {
-      recordRequest(supported({ read: i, total: 10 }), `m${i}`)
-    }
-    const value = await runCommand()
-    expect(value).not.toContain('older omitted')
-    expect(value).toContain('(5)')
-  })
-})
-
-describe('/cache-stats — model label rendering', () => {
-  test('truncates long model labels to fit the column width', async () => {
-    // cacheStats.ts pads+slices the label to 28 chars for alignment.
-    const longLabel = 'some-extremely-long-model-identifier-that-wraps'
-    recordRequest(supported({ read: 10, total: 100, hitRate: 0.1 }), longLabel)
-    const value = await runCommand()
-    // Sliced to 28 chars.
-    expect(value).toContain(longLabel.slice(0, 28))
-    // And the full string should NOT appear (would mean no truncation).
-    expect(value).not.toContain(longLabel)
-  })
-})
-
-describe('/cache-stats — timestamp rendering', () => {
-  test('renders each row with full date and time (YYYY-MM-DD HH:MM:SS)', async () => {
-    recordRequest(supported({ read: 5, total: 10, hitRate: 0.5 }), 'claude-x')
-    const value = await runCommand()
-    // Match the full ISO-ish date + time the row uses. We assert the shape,
-    // not a specific timestamp — real clock is used, so a regex on the
-    // format is the right assertion.
-    expect(value).toMatch(/\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}/)
-    // Bare time-of-day alone (no date) should NOT appear in isolation — it
-    // must always be preceded by the date. Guards against regression if
-    // someone shortens the formatter again.
-    const timeOnlyInRow = /\n\s*#\s*\d+\s+\d{2}:\d{2}:\d{2}\s/.test(value)
-    expect(timeOnlyInRow).toBe(false)
-  })
-})

src/commands/cacheStats/cacheStats.ts

@@ -1,74 +0,0 @@
-import {
-  getCacheStatsHistory,
-  getCurrentTurnCacheMetrics,
-  getSessionCacheMetrics,
-  type CacheStatsEntry,
-} from '../../services/api/cacheStatsTracker.js'
-import {
-  formatCacheMetricsCompact,
-  formatCacheMetricsFull,
-  type CacheMetrics,
-} from '../../services/api/cacheMetrics.js'
-import type { LocalCommandCall } from '../../types/command.js'
-
-// Cap the per-request breakdown to keep output readable. Users wanting
-// the full history can rely on OPENCLAUDE_LOG_TOKEN_USAGE=verbose for
-// structured per-request stderr output.
-const MAX_RECENT_ROWS = 20
-
-function formatRow(entry: CacheStatsEntry, idx: number): string {
-  // `YYYY-MM-DD HH:MM:SS` — long-running sessions can span midnight and a
-  // bare time-of-day makes the wrong row look "most recent" when two
-  // entries on different days share the same HH:MM:SS.
-  const iso = new Date(entry.timestamp).toISOString()
-  const ts = `${iso.slice(0, 10)} ${iso.slice(11, 19)}`
-  const line = formatCacheMetricsCompact(entry.metrics)
-  return `  #${String(idx + 1).padStart(3)}  ${ts}  ${entry.label.padEnd(28).slice(0, 28)}  ${line}`
-}
-
-function summarize(label: string, m: CacheMetrics): string {
-  return `${label.padEnd(18)}${formatCacheMetricsFull(m)}`
-}
-
-export const call: LocalCommandCall = async () => {
-  const history = getCacheStatsHistory()
-  const session = getSessionCacheMetrics()
-  const turn = getCurrentTurnCacheMetrics()
-
-  if (history.length === 0) {
-    return {
-      type: 'text',
-      value:
-        'Cache stats\n  No API requests yet this session.\n  Start a turn and re-run /cache-stats to see results.',
-    }
-  }
-
-  const recent = history.slice(-MAX_RECENT_ROWS)
-  const omitted = history.length - recent.length
-
-  const lines: string[] = ['Cache stats', '']
-  lines.push(summarize('Current turn:', turn))
-  lines.push(summarize('Session total:', session))
-  lines.push('')
-  lines.push(`Recent requests (${recent.length}${omitted > 0 ? ` of ${history.length}, ${omitted} older omitted` : ''}):`)
-  lines.push(`  #     time      model                         cache`)
-  for (const [i, entry] of recent.entries()) {
-    lines.push(formatRow(entry, history.length - recent.length + i))
-  }
-
-  // Honesty footnote — providers without cache reporting (vanilla Copilot,
-  // Ollama) show [Cache: N/A] rather than a fake 0%. Tell the user so they
-  // don't read "N/A" as "broken".
-  const hasUnsupported = recent.some((e) => !e.metrics.supported)
-  if (hasUnsupported) {
-    lines.push('')
-    lines.push(
-      '  N/A rows: provider API does not expose cache usage (GitHub Copilot, Ollama).',
-    )
-    lines.push(
-      '  The request still ran normally — only the metric is unavailable.',
-    )
-  }
-
-  return { type: 'text', value: lines.join('\n') }
-}

src/commands/cacheStats/index.ts

@@ -1,24 +0,0 @@
-/**
- * /cache-stats — per-session cache diagnostics.
- *
- * Always-on diagnostic command (no toggle) that surfaces the metrics
- * tracked in `cacheStatsTracker.ts`. Breaks cache usage down by request
- * and also reports the session-wide aggregate — useful when the user
- * suspects a cache bust (e.g. after /reload-plugins) and wants to see
- * whether recent turns still hit the cache.
- *
- * Lazy-loaded (implementation in cacheStats.ts) to keep startup time
- * minimal — same pattern used by /cost and /cache-probe.
- */
-import type { Command } from '../../commands.js'
-
-const cacheStats = {
-  type: 'local',
-  name: 'cache-stats',
-  description:
-    'Show per-turn and session cache hit/miss stats (works across all providers)',
-  supportsNonInteractive: true,
-  load: () => import('./cacheStats.js'),
-} satisfies Command
-
-export default cacheStats

src/commands/chrome/chrome.tsx

@@ -197,7 +197,7 @@ function ClaudeInChromeMenu(t0) {
   }
   let t6;
   if ($[20] === Symbol.for("react.memo_cache_sentinel")) {
-    t6 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from OpenClaude. Navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.</Text>;
+    t6 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from Claude Code. Navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.</Text>;
     $[20] = t6;
   } else {
     t6 = $[20];

src/commands/createMovedToPluginCommand.ts

@@ -48,7 +48,7 @@ export function createMovedToPluginCommand({
             text: `This command has been moved to a plugin. Tell the user:
 
 1. To install the plugin, run:
-   openclaude plugin install ${pluginName}@claude-code-marketplace
+   claude plugin install ${pluginName}@claude-code-marketplace
 
 2. After installation, use /${pluginName}:${pluginCommand} to run this command
 

src/commands/doctor/index.ts

@@ -3,7 +3,7 @@ import { isEnvTruthy } from '../../utils/envUtils.js'
 
 const doctor: Command = {
   name: 'doctor',
-  description: 'Diagnose and verify your OpenClaude installation and settings',
+  description: 'Diagnose and verify your Claude Code installation and settings',
   isEnabled: () => !isEnvTruthy(process.env.DISABLE_DOCTOR_COMMAND),
   type: 'local-jsx',
   load: () => import('./doctor.js'),

src/commands/feedback/index.ts

@@ -7,7 +7,7 @@ const feedback = {
   aliases: ['bug'],
   type: 'local-jsx',
   name: 'feedback',
-  description: `Submit feedback about OpenClaude`,
+  description: `Submit feedback about Claude Code`,
   argumentHint: '[report]',
   isEnabled: () =>
     !(

src/commands/insights.ts

@@ -247,7 +247,7 @@ function getSessionMetaDir(): string {
   return join(getDataDir(), 'session-meta')
 }
 
-const FACET_EXTRACTION_PROMPT = `Analyze this OpenClaude session and extract structured facets.
+const FACET_EXTRACTION_PROMPT = `Analyze this Claude Code session and extract structured facets.
 
 CRITICAL GUIDELINES:
 
@@ -687,7 +687,7 @@ function formatTranscriptForFacets(log: LogOption): string {
   return lines.join('\n')
 }
 
-const SUMMARIZE_CHUNK_PROMPT = `Summarize this portion of a OpenClaude session transcript. Focus on:
+const SUMMARIZE_CHUNK_PROMPT = `Summarize this portion of a Claude Code session transcript. Focus on:
 1. What the user asked for
 2. What Claude did (tools used, files modified)
 3. Any friction or issues
@@ -1156,12 +1156,12 @@ type InsightSection = {
 const INSIGHT_SECTIONS: InsightSection[] = [
   {
     name: 'project_areas',
-    prompt: `Analyze this OpenClaude usage data and identify project areas.
+    prompt: `Analyze this Claude Code usage data and identify project areas.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
   "areas": [
-    {"name": "Area name", "session_count": N, "description": "2-3 sentences about what was worked on and how OpenClaude was used."}
+    {"name": "Area name", "session_count": N, "description": "2-3 sentences about what was worked on and how Claude Code was used."}
   ]
 }
 
@@ -1170,18 +1170,18 @@ Include 4-5 areas. Skip internal CC operations.`,
   },
   {
     name: 'interaction_style',
-    prompt: `Analyze this OpenClaude usage data and describe the user's interaction style.
+    prompt: `Analyze this Claude Code usage data and describe the user's interaction style.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
-  "narrative": "2-3 paragraphs analyzing HOW the user interacts with OpenClaude. Use second person 'you'. Describe patterns: iterate quickly vs detailed upfront specs? Interrupt often or let Claude run? Include specific examples. Use **bold** for key insights.",
+  "narrative": "2-3 paragraphs analyzing HOW the user interacts with Claude Code. Use second person 'you'. Describe patterns: iterate quickly vs detailed upfront specs? Interrupt often or let Claude run? Include specific examples. Use **bold** for key insights.",
   "key_pattern": "One sentence summary of most distinctive interaction style"
 }`,
     maxTokens: 8192,
   },
   {
     name: 'what_works',
-    prompt: `Analyze this OpenClaude usage data and identify what's working well for this user. Use second person ("you").
+    prompt: `Analyze this Claude Code usage data and identify what's working well for this user. Use second person ("you").
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1196,7 +1196,7 @@ Include 3 impressive workflows.`,
   },
   {
     name: 'friction_analysis',
-    prompt: `Analyze this OpenClaude usage data and identify friction points for this user. Use second person ("you").
+    prompt: `Analyze this Claude Code usage data and identify friction points for this user. Use second person ("you").
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1211,7 +1211,7 @@ Include 3 friction categories with 2 examples each.`,
   },
   {
     name: 'suggestions',
-    prompt: `Analyze this OpenClaude usage data and suggest improvements.
+    prompt: `Analyze this Claude Code usage data and suggest improvements.
 
 ## CC FEATURES REFERENCE (pick from these for features_to_try):
 1. **MCP Servers**: Connect Claude to external tools, databases, and APIs via Model Context Protocol.
@@ -1254,7 +1254,7 @@ IMPORTANT for features_to_try: Pick 2-3 from the CC FEATURES REFERENCE above. In
   },
   {
     name: 'on_the_horizon',
-    prompt: `Analyze this OpenClaude usage data and identify future opportunities.
+    prompt: `Analyze this Claude Code usage data and identify future opportunities.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1271,7 +1271,7 @@ Include 3 opportunities. Think BIG - autonomous workflows, parallel agents, iter
     ? [
         {
           name: 'cc_team_improvements',
-          prompt: `Analyze this OpenClaude usage data and suggest product improvements for the CC team.
+          prompt: `Analyze this Claude Code usage data and suggest product improvements for the CC team.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1285,7 +1285,7 @@ Include 2-3 improvements based on friction patterns observed.`,
         },
         {
           name: 'model_behavior_improvements',
-          prompt: `Analyze this OpenClaude usage data and suggest model behavior improvements.
+          prompt: `Analyze this Claude Code usage data and suggest model behavior improvements.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1301,7 +1301,7 @@ Include 2-3 improvements based on friction patterns observed.`,
     : []),
   {
     name: 'fun_ending',
-    prompt: `Analyze this OpenClaude usage data and find a memorable moment.
+    prompt: `Analyze this Claude Code usage data and find a memorable moment.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1555,7 +1555,7 @@ async function generateParallelInsights(
       .join('\n') || ''
 
   // Now generate "At a Glance" with access to other sections' outputs
-  const atAGlancePrompt = `You're writing an "At a Glance" summary for a OpenClaude usage insights report for OpenClaude users. The goal is to help them understand their usage and improve how they can use Claude better, especially as models improve.
+  const atAGlancePrompt = `You're writing an "At a Glance" summary for a Claude Code usage insights report for Claude Code users. The goal is to help them understand their usage and improve how they can use Claude better, especially as models improve.
 
 Use this 4-part structure:
 
@@ -1563,7 +1563,7 @@ Use this 4-part structure:
 
 2. **What's hindering you** - Split into (a) Claude's fault (misunderstandings, wrong approaches, bugs) and (b) user-side friction (not providing enough context, environment issues -- ideally more general than just one project). Be honest but constructive.
 
-3. **Quick wins to try** - Specific OpenClaude features they could try from the examples below, or a workflow technique if you think it's really compelling. (Avoid stuff like "Ask Claude to confirm before taking actions" or "Type out more context up front" which are less compelling.)
+3. **Quick wins to try** - Specific Claude Code features they could try from the examples below, or a workflow technique if you think it's really compelling. (Avoid stuff like "Ask Claude to confirm before taking actions" or "Type out more context up front" which are less compelling.)
 
 4. **Ambitious workflows for better models** - As we move to much more capable models over the next 3-6 months, what should they prepare for? What workflows that seem impossible now will become possible? Draw from the appropriate section below.
 
@@ -1826,7 +1826,7 @@ function generateHtmlReport(
   const interactionStyle = insights.interaction_style
   const interactionHtml = interactionStyle?.narrative
     ? `
-    <h2 id="section-usage">How You Use OpenClaude</h2>
+    <h2 id="section-usage">How You Use Claude Code</h2>
     <div class="narrative">
       ${markdownToHtml(interactionStyle.narrative)}
       ${interactionStyle.key_pattern ? `<div class="key-insight"><strong>Key pattern:</strong> ${escapeHtml(interactionStyle.key_pattern)}</div>` : ''}
@@ -1890,7 +1890,7 @@ function generateHtmlReport(
     <h2 id="section-features">Existing CC Features to Try</h2>
     <div class="claude-md-section">
       <h3>Suggested CLAUDE.md Additions</h3>
-      <p style="font-size: 12px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude to add it to your CLAUDE.md.</p>
+      <p style="font-size: 12px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code to add it to your CLAUDE.md.</p>
       <div class="claude-md-actions">
         <button class="copy-all-btn" onclick="copyAllCheckedClaudeMd()">Copy All Checked</button>
       </div>
@@ -1915,7 +1915,7 @@ function generateHtmlReport(
     ${
       suggestions.features_to_try && suggestions.features_to_try.length > 0
         ? `
-    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude and it'll set it up for you.</p>
+    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code and it'll set it up for you.</p>
     <div class="features-section">
       ${suggestions.features_to_try
         .map(
@@ -1949,8 +1949,8 @@ function generateHtmlReport(
     ${
       suggestions.usage_patterns && suggestions.usage_patterns.length > 0
         ? `
-    <h2 id="section-patterns">New Ways to Use OpenClaude</h2>
-    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude and it'll walk you through it.</p>
+    <h2 id="section-patterns">New Ways to Use Claude Code</h2>
+    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code and it'll walk you through it.</p>
     <div class="patterns-section">
       ${suggestions.usage_patterns
         .map(
@@ -1963,7 +1963,7 @@ function generateHtmlReport(
             pat.copyable_prompt
               ? `
           <div class="copyable-prompt-section">
-            <div class="prompt-label">Paste into OpenClaude:</div>
+            <div class="prompt-label">Paste into Claude Code:</div>
             <div class="copyable-prompt-row">
               <code class="copyable-prompt">${escapeHtml(pat.copyable_prompt)}</code>
               <button class="copy-btn" onclick="copyText(this)">Copy</button>
@@ -1998,7 +1998,7 @@ function generateHtmlReport(
           <div class="horizon-title">${escapeHtml(opp.title || '')}</div>
           <div class="horizon-possible">${escapeHtml(opp.whats_possible || '')}</div>
           ${opp.how_to_try ? `<div class="horizon-tip"><strong>Getting started:</strong> ${escapeHtml(opp.how_to_try)}</div>` : ''}
-          ${opp.copyable_prompt ? `<div class="pattern-prompt"><div class="prompt-label">Paste into OpenClaude:</div><code>${escapeHtml(opp.copyable_prompt)}</code><button class="copy-btn" onclick="copyText(this)">Copy</button></div>` : ''}
+          ${opp.copyable_prompt ? `<div class="pattern-prompt"><div class="prompt-label">Paste into Claude Code:</div><code>${escapeHtml(opp.copyable_prompt)}</code><button class="copy-btn" onclick="copyText(this)">Copy</button></div>` : ''}
         </div>
       `,
         )
@@ -2305,13 +2305,13 @@ function generateHtmlReport(
 <html>
 <head>
   <meta charset="utf-8">
-  <title>OpenClaude Insights</title>
+  <title>Claude Code Insights</title>
   <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap" rel="stylesheet">
   <style>${css}</style>
 </head>
 <body>
   <div class="container">
-    <h1>OpenClaude Insights</h1>
+    <h1>Claude Code Insights</h1>
     <p class="subtitle">${data.total_messages.toLocaleString()} messages across ${data.total_sessions} sessions${data.total_sessions_scanned && data.total_sessions_scanned > data.total_sessions ? ` (${data.total_sessions_scanned.toLocaleString()} total)` : ''} | ${data.date_range.start} to ${data.date_range.end}</p>
 
     ${atAGlanceHtml}
@@ -2377,7 +2377,7 @@ function generateHtmlReport(
         data.multi_clauding.overlap_events === 0
           ? `
         <p style="font-size: 14px; color: #64748b; padding: 8px 0;">
-          No parallel session usage detected. You typically work with one OpenClaude session at a time.
+          No parallel session usage detected. You typically work with one Claude Code session at a time.
         </p>
       `
           : `
@@ -2396,7 +2396,7 @@ function generateHtmlReport(
           </div>
         </div>
         <p style="font-size: 13px; color: #475569; margin-top: 12px;">
-          You run multiple OpenClaude sessions simultaneously. Multi-clauding is detected when sessions
+          You run multiple Claude Code sessions simultaneously. Multi-clauding is detected when sessions
           overlap in time, suggesting parallel workflows.
         </p>
       `
@@ -2836,7 +2836,7 @@ function safeKeys(obj: Record<string, unknown> | undefined | null): string[] {
 const usageReport: Command = {
   type: 'prompt',
   name: 'insights',
-  description: 'Generate a report analyzing your OpenClaude sessions',
+  description: 'Generate a report analyzing your Claude Code sessions',
   contentLength: 0, // Dynamic content
   progressMessage: 'analyzing your sessions',
   source: 'builtin',
@@ -2874,7 +2874,7 @@ ${atAGlance.quick_wins ? `**Quick wins to try:** ${atAGlance.quick_wins} See _Fe
 ${atAGlance.ambitious_workflows ? `**Ambitious workflows:** ${atAGlance.ambitious_workflows} See _On the Horizon_.` : ''}`
       : '_No insights generated_'
 
-    const header = `# OpenClaude Insights
+    const header = `# Claude Code Insights
 
 ${stats}
 ${data.date_range.start} to ${data.date_range.end}
@@ -2888,7 +2888,7 @@ Your full shareable insights report is ready: ${reportUrl}${uploadHint}`
     return [
       {
         type: 'text',
-        text: `The user just ran /insights to generate a usage report analyzing their OpenClaude sessions.
+        text: `The user just ran /insights to generate a usage report analyzing their Claude Code sessions.
 
 Here is the full insights data:
 ${jsonStringify(insights, null, 2)}

src/commands/install.tsx

@@ -210,12 +210,12 @@ function Install({
   useEffect(() => {
     if (state.type === 'success') {
       // Give success message time to render before exiting
-      setTimeout(onDone, 2000, 'OpenClaude installation completed successfully', {
+      setTimeout(onDone, 2000, 'Claude Code installation completed successfully', {
         display: 'system' as const
       });
     } else if (state.type === 'error') {
       // Give error message time to render before exiting
-      setTimeout(onDone, 3000, 'OpenClaude installation failed', {
+      setTimeout(onDone, 3000, 'Claude Code installation failed', {
         display: 'system' as const
       });
     }
@@ -226,7 +226,7 @@ function Install({
       {state.type === 'cleaning-npm' && <Text color="warning">Cleaning up old npm installations...</Text>}
 
       {state.type === 'installing' && <Text color="claude">
-          Installing OpenClaude native build {state.version}...
+          Installing Claude Code native build {state.version}...
         </Text>}
 
       {state.type === 'setting-up' && <Text color="claude">Setting up launcher and shell integration...</Text>}
@@ -237,7 +237,7 @@ function Install({
           <Box>
             <StatusIcon status="success" withSpace />
             <Text color="success" bold>
-              OpenClaude successfully installed!
+              Claude Code successfully installed!
             </Text>
           </Box>
           <Box marginLeft={2} flexDirection="column" gap={1}>
@@ -254,7 +254,7 @@ function Install({
             <Box marginTop={1}>
               <Text dimColor>Next: Run </Text>
               <Text color="claude" bold>
-                openclaude --help
+                claude --help
               </Text>
               <Text dimColor> to get started</Text>
             </Box>
@@ -279,7 +279,7 @@ function Install({
 export const install = {
   type: 'local-jsx' as const,
   name: 'install',
-  description: 'Install OpenClaude native build',
+  description: 'Install Claude Code native build',
   argumentHint: '[options]',
   async call(onDone: (result: string, options?: {
     display?: CommandResultDisplay;

src/commands/knowledge/index.ts

@@ -1,12 +0,0 @@
-import type { Command } from '../../commands.js'
-
-const knowledge: Command = {
-  type: 'local',
-  name: 'knowledge',
-  description: 'Manage native Knowledge Graph',
-  supportsNonInteractive: true,
-  argumentHint: 'enable <yes|no> | clear | status | list',
-  load: () => import('./knowledge.js'),
-}
-
-export default knowledge

src/commands/knowledge/knowledge.test.ts

@@ -1,74 +0,0 @@
-import { describe, expect, it, beforeEach } from 'bun:test'
-import { call as knowledgeCall } from './knowledge.js'
-import { getGlobalConfig, saveGlobalConfig } from '../../utils/config.js'
-import { getArc, addEntity, resetArc } from '../../utils/conversationArc.js'
-import { getGlobalGraph, resetGlobalGraph } from '../../utils/knowledgeGraph.js'
-
-describe('knowledge command', () => {
-  const mockContext = {} as any
-
-  beforeEach(() => {
-    resetArc()
-    resetGlobalGraph()
-  })
-  
-  const knowledgeCallWithCapture = async (args: string) => {
-    const result = await knowledgeCall(args, mockContext)
-    if (result.type === 'text') {
-      return result.value
-    }
-    return ''
-  }
-
-  beforeEach(() => {
-    // Attempt to reset config - even if mocked, we try to set our key
-    try {
-      saveGlobalConfig(current => ({
-        ...current,
-        knowledgeGraphEnabled: true
-      }))
-    } catch {
-      // Ignore if config is heavily mocked
-    }
-    resetArc()
-  })
-
-  it('enables and disables knowledge graph engine', async () => {
-    // Test Disable
-    const res1 = await knowledgeCallWithCapture('enable no')
-    expect(res1.toLowerCase()).toContain('disabled')
-    
-    // Safety check: only verify state if property is actually present (avoid CI mock interference)
-    const config1 = getGlobalConfig()
-    if (config1 && 'knowledgeGraphEnabled' in config1) {
-      expect(config1.knowledgeGraphEnabled).toBe(false)
-    }
-
-    // Test Enable
-    const res2 = await knowledgeCallWithCapture('enable yes')
-    expect(res2.toLowerCase()).toContain('enabled')
-    
-    const config2 = getGlobalConfig()
-    if (config2 && 'knowledgeGraphEnabled' in config2) {
-      expect(config2.knowledgeGraphEnabled).toBe(true)
-    }
-  })
-
-  it('clears the knowledge graph', async () => {
-    // Add a fact first
-    addEntity('test', 'fact')
-    const graph = getGlobalGraph()
-    expect(Object.keys(graph.entities).length).toBe(1)
-
-    // Clear it
-    const res = await knowledgeCallWithCapture('clear')
-    const graphAfter = getGlobalGraph()
-    expect(Object.keys(graphAfter.entities).length).toBe(0)
-    expect(res.toLowerCase()).toContain('cleared')
-  })
-
-  it('shows error on unknown subcommand', async () => {
-    const res = await knowledgeCallWithCapture('invalid')
-    expect(res.toLowerCase()).toContain('unknown subcommand')
-  })
-})

src/commands/knowledge/knowledge.ts

@@ -1,63 +0,0 @@
-import type { LocalCommandCall } from '../../types/command.js';
-import { getArcSummary, resetArc, getArcStats } from '../../utils/conversationArc.js';
-import { getGlobalGraph, resetGlobalGraph } from '../../utils/knowledgeGraph.js';
-import { getGlobalConfig, saveGlobalConfig } from '../../utils/config.js';
-import chalk from 'chalk';
-
-export const call: LocalCommandCall = async (args, _context) => {
-  const arg = (args ? String(args) : '').trim().toLowerCase();
-  const splitArgs = arg.split(/\s+/).filter(Boolean);
-  const subCommand = splitArgs[0];
-
-  if (!subCommand || subCommand === 'status') {
-    const config = getGlobalConfig();
-    const stats = getArcStats();
-    const graph = getGlobalGraph();
-    const entityCount = Object.keys(graph.entities).length;
-    
-    const statusText = (config.knowledgeGraphEnabled !== false)
-      ? chalk.green('ENABLED') 
-      : chalk.red('DISABLED');
-      
-    let output = `${chalk.bold('Knowledge Graph Engine')}: ${statusText}\n`;
-    if (stats) {
-      output += `• Stats: ${stats.goalCount} goals, ${stats.milestoneCount} milestones, ${entityCount} technical facts learned`;
-    }
-    
-    return { type: 'text', value: output };
-  }
-
-  if (subCommand === 'enable') {
-    const val = splitArgs[1];
-    const isEnabled = val === 'yes' || val === 'true';
-    const isDisabled = val === 'no' || val === 'false';
-
-    if (!isEnabled && !isDisabled) {
-      return { type: 'text', value: 'Usage: /knowledge enable <yes|no>' };
-    }
-
-    saveGlobalConfig(current => ({ ...current, knowledgeGraphEnabled: isEnabled }));
-    return { 
-      type: 'text', 
-      value: `✨ Knowledge Graph engine ${isEnabled ? chalk.green('enabled') : chalk.red('disabled')}.` 
-    };
-  }
-
-  if (subCommand === 'clear') {
-    resetArc();
-    resetGlobalGraph();
-    return { 
-      type: 'text', 
-      value: '🗑️ Knowledge graph memory has been cleared for this session.' 
-    };
-  }
-
-  if (subCommand === 'list') {
-    return { type: 'text', value: getArcSummary() };
-  }
-
-  return { 
-    type: 'text', 
-    value: `Unknown subcommand: ${subCommand}. Available: enable, clear, status, list` 
-  };
-};

src/commands/mcp/addCommand.ts

@@ -34,16 +34,16 @@ export function registerMcpAddCommand(mcp: Command): void {
   mcp
     .command('add <name> <commandOrUrl> [args...]')
     .description(
-      'Add an MCP server to OpenClaude.\n\n' +
+      'Add an MCP server to Claude Code.\n\n' +
         'Examples:\n' +
         '  # Add HTTP server:\n' +
-        '  openclaude mcp add --transport http sentry https://mcp.sentry.dev/mcp\n\n' +
+        '  claude mcp add --transport http sentry https://mcp.sentry.dev/mcp\n\n' +
         '  # Add HTTP server with headers:\n' +
-        '  openclaude mcp add --transport http corridor https://app.corridor.dev/api/mcp --header "Authorization: Bearer ..."\n\n' +
+        '  claude mcp add --transport http corridor https://app.corridor.dev/api/mcp --header "Authorization: Bearer ..."\n\n' +
         '  # Add stdio server with environment variables:\n' +
-        '  openclaude mcp add -e API_KEY=xxx my-server -- npx my-mcp-server\n\n' +
+        '  claude mcp add -e API_KEY=xxx my-server -- npx my-mcp-server\n\n' +
         '  # Add stdio server with subprocess flags:\n' +
-        '  openclaude mcp add my-server -- my-command --some-flag arg1',
+        '  claude mcp add my-server -- my-command --some-flag arg1',
     )
     .option(
       '-s, --scope <scope>',
@@ -75,7 +75,7 @@ export function registerMcpAddCommand(mcp: Command): void {
     .addOption(
       new Option(
         '--xaa',
-        "Enable XAA (SEP-990) for this server. Requires 'openclaude mcp xaa setup' first. Also requires --client-id and --client-secret (for the MCP server's AS).",
+        "Enable XAA (SEP-990) for this server. Requires 'claude mcp xaa setup' first. Also requires --client-id and --client-secret (for the MCP server's AS).",
       ).hideHelp(!isXaaEnabled()),
     )
     .action(async (name, commandOrUrl, args, options) => {
@@ -87,12 +87,12 @@ export function registerMcpAddCommand(mcp: Command): void {
       if (!name) {
         cliError(
           'Error: Server name is required.\n' +
-            'Usage: openclaude mcp add <name> <command> [args...]',
+            'Usage: claude mcp add <name> <command> [args...]',
         )
       } else if (!actualCommand) {
         cliError(
           'Error: Command is required when server name is provided.\n' +
-            'Usage: openclaude mcp add <name> <command> [args...]',
+            'Usage: claude mcp add <name> <command> [args...]',
         )
       }
 
@@ -113,7 +113,7 @@ export function registerMcpAddCommand(mcp: Command): void {
           if (!options.clientSecret) missing.push('--client-secret')
           if (!getXaaIdpSettings()) {
             missing.push(
-              "'openclaude mcp xaa setup' (settings.xaaIdp not configured)",
+              "'claude mcp xaa setup' (settings.xaaIdp not configured)",
             )
           }
           if (missing.length) {
@@ -254,10 +254,10 @@ export function registerMcpAddCommand(mcp: Command): void {
               `\nWarning: The command "${actualCommand}" looks like a URL, but is being interpreted as a stdio server as --transport was not specified.\n`,
             )
             process.stderr.write(
-              `If this is an HTTP server, use: openclaude mcp add --transport http ${name} ${actualCommand}\n`,
+              `If this is an HTTP server, use: claude mcp add --transport http ${name} ${actualCommand}\n`,
             )
             process.stderr.write(
-              `If this is an SSE server, use: openclaude mcp add --transport sse ${name} ${actualCommand}\n`,
+              `If this is an SSE server, use: claude mcp add --transport sse ${name} ${actualCommand}\n`,
             )
           }
 

src/commands/mcp/xaaIdpCommand.ts

@@ -170,7 +170,7 @@ export function registerMcpXaaIdpCommand(mcp: Command): void {
       const idp = getXaaIdpSettings()
       if (!idp) {
         return cliError(
-          "Error: no XAA IdP connection. Run 'openclaude mcp xaa setup' first.",
+          "Error: no XAA IdP connection. Run 'claude mcp xaa setup' first.",
         )
       }
 
@@ -235,7 +235,7 @@ export function registerMcpXaaIdpCommand(mcp: Command): void {
         `Client secret: ${hasSecret ? '(stored in keychain)' : '(not set — PKCE-only)'}\n`,
       )
       process.stdout.write(
-        `Logged in:     ${hasIdToken ? 'yes (id_token cached)' : "no — run 'openclaude mcp xaa login'"}\n`,
+        `Logged in:     ${hasIdToken ? 'yes (id_token cached)' : "no — run 'claude mcp xaa login'"}\n`,
       )
       cliOk()
     })

src/commands/model/index.ts

@@ -6,7 +6,7 @@ export default {
   type: 'local-jsx',
   name: 'model',
   get description() {
-    return `Set the AI model for OpenClaude (currently ${renderModelName(getMainLoopModel())})`
+    return `Set the AI model for Claude Code (currently ${renderModelName(getMainLoopModel())})`
   },
   argumentHint: '[model]',
   get immediate() {

src/commands/plugin/DiscoverPlugins.tsx

@@ -713,7 +713,7 @@ function EmptyStateMessage(t0) {
       {
         let t1;
         if ($[0] === Symbol.for("react.memo_cache_sentinel")) {
-          t1 = <><Text dimColor={true}>Git is required to install marketplaces.</Text><Text dimColor={true}>Please install git and restart OpenClaude.</Text></>;
+          t1 = <><Text dimColor={true}>Git is required to install marketplaces.</Text><Text dimColor={true}>Please install git and restart Claude Code.</Text></>;
           $[0] = t1;
         } else {
           t1 = $[0];

src/commands/plugin/index.tsx

@@ -3,7 +3,7 @@ const plugin = {
   type: 'local-jsx',
   name: 'plugin',
   aliases: ['plugins', 'marketplace'],
-  description: 'Manage OpenClaude plugins',
+  description: 'Manage Claude Code plugins',
   immediate: true,
   load: () => import('./plugin.js')
 } satisfies Command;

src/commands/provider/provider.test.tsx

@@ -11,7 +11,6 @@ import {
   buildCodexOAuthProfileEnv,
   buildCurrentProviderSummary,
   buildProfileSaveMessage,
-  buildProviderManagerCompletion,
   getProviderWizardDefaults,
   ProviderWizard,
   TextEntryDialog,
@@ -265,32 +264,6 @@ test('wizard step remount prevents a typed API key from leaking into the next fi
   expect(output).not.toContain('sk-secret-12345678')
 })
 
-test('buildProviderManagerCompletion records provider switch event and model-visible reminder', () => {
-  const completion = buildProviderManagerCompletion({
-    action: 'activated',
-    activeProviderName: 'Sadaf Provider',
-    activeProviderModel: 'sadaf-model',
-    message: 'Provider switched to Sadaf Provider (sadaf-model)',
-  })
-
-  expect(completion.message).toBe(
-    'Provider switched to Sadaf Provider (sadaf-model)',
-  )
-  expect(completion.metaMessages).toEqual([
-    '<system-reminder>Provider switched mid-session to Sadaf Provider using model sadaf-model. Use this provider/model for subsequent requests unless the user switches again.</system-reminder>',
-  ])
-})
-
-test('buildProviderManagerCompletion skips provider reminder when manager is cancelled', () => {
-  const completion = buildProviderManagerCompletion({
-    action: 'cancelled',
-    message: 'Provider manager closed',
-  })
-
-  expect(completion.message).toBe('Provider manager closed')
-  expect(completion.metaMessages).toBeUndefined()
-})
-
 test('buildProfileSaveMessage maps provider fields without echoing secrets', () => {
   const message = buildProfileSaveMessage(
     'openai',

src/commands/provider/provider.tsx

@@ -2,10 +2,7 @@ import * as React from 'react'
 
 import type { LocalJSXCommandCall, LocalJSXCommandOnDone } from '../../types/command.js'
 import { COMMON_HELP_ARGS, COMMON_INFO_ARGS } from '../../constants/xml.js'
-import {
-  ProviderManager,
-  type ProviderManagerResult,
-} from '../../components/ProviderManager.js'
+import { ProviderManager } from '../../components/ProviderManager.js'
 import TextInput from '../../components/TextInput.js'
 import {
   Select,
@@ -73,29 +70,6 @@ import {
   type OllamaGenerationReadiness,
 } from '../../utils/providerDiscovery.js'
 
-export function buildProviderManagerCompletion(result?: ProviderManagerResult): {
-  message: string
-  metaMessages?: string[]
-} {
-  const message =
-    result?.message ??
-    (result?.action === 'saved'
-      ? 'Provider profile updated'
-      : 'Provider manager closed')
-  const metaMessages =
-    result?.action === 'activated' && result.activeProviderName
-      ? [
-          `<system-reminder>Provider switched mid-session to ${result.activeProviderName}${
-            result.activeProviderModel
-              ? ` using model ${result.activeProviderModel}`
-              : ''
-          }. Use this provider/model for subsequent requests unless the user switches again.</system-reminder>`,
-        ]
-      : undefined
-
-  return { message, metaMessages }
-}
-
 function describeOllamaReadinessIssue(
   readiness: OllamaGenerationReadiness,
   options?: {
@@ -1729,8 +1703,13 @@ export const call: LocalJSXCommandCall = async (onDone, _context, args) => {
     <ProviderManager
       mode="manage"
       onDone={result => {
-        const { message, metaMessages } = buildProviderManagerCompletion(result)
-        onDone(message, { display: 'system', metaMessages })
+        const message =
+          result?.message ??
+          (result?.action === 'saved'
+            ? 'Provider profile updated'
+            : 'Provider manager closed')
+
+        onDone(message, { display: 'system' })
       }}
     />
   )

src/commands/remote-setup/index.ts

@@ -6,7 +6,7 @@ const web = {
   type: 'local-jsx',
   name: 'web-setup',
   description:
-    'Setup OpenClaude on the web (requires connecting your GitHub account)',
+    'Setup Claude Code on the web (requires connecting your GitHub account)',
   availability: ['claude-ai'],
   isEnabled: () =>
     getFeatureValue_CACHED_MAY_BE_STALE('tengu_cobalt_lantern', false) &&

src/commands/review.ts

@@ -48,7 +48,7 @@ const review: Command = {
 const ultrareview: Command = {
   type: 'local-jsx',
   name: 'ultrareview',
-  description: `~10–20 min · Finds and verifies bugs in your branch. Runs in OpenClaude on the web. See ${CCR_TERMS_URL}`,
+  description: `~10–20 min · Finds and verifies bugs in your branch. Runs in Claude Code on the web. See ${CCR_TERMS_URL}`,
   isEnabled: () => isUltrareviewEnabled(),
   load: () => import('./review/ultrareviewCommand.js'),
 }

src/commands/session/session.tsx

@@ -57,7 +57,7 @@ function SessionInfo(t0) {
   if (!remoteSessionUrl) {
     let t4;
     if ($[4] === Symbol.for("react.memo_cache_sentinel")) {
-      t4 = <Pane><Text color="warning">Not in remote mode. Start with `openclaude --remote` to use this command.</Text><Text dimColor={true}>(press esc to close)</Text></Pane>;
+      t4 = <Pane><Text color="warning">Not in remote mode. Start with `claude --remote` to use this command.</Text><Text dimColor={true}>(press esc to close)</Text></Pane>;
       $[4] = t4;
     } else {
       t4 = $[4];

src/commands/stats/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const stats = {
   type: 'local-jsx',
   name: 'stats',
-  description: 'Show your OpenClaude usage statistics and activity',
+  description: 'Show your Claude Code usage statistics and activity',
   load: () => import('./stats.js'),
 } satisfies Command
 

src/commands/status/index.ts

@@ -4,7 +4,7 @@ const status = {
   type: 'local-jsx',
   name: 'status',
   description:
-    'Show OpenClaude status including version, model, account, API connectivity, and tool statuses',
+    'Show Claude Code status including version, model, account, API connectivity, and tool statuses',
   immediate: true,
   load: () => import('./status.js'),
 } satisfies Command

src/commands/statusline.tsx

@@ -3,7 +3,7 @@ import type { Command } from '../commands.js';
 import { AGENT_TOOL_NAME } from '../tools/AgentTool/constants.js';
 const statusline = {
   type: 'prompt',
-  description: "Set up OpenClaude's status line UI",
+  description: "Set up Claude Code's status line UI",
   contentLength: 0,
   // Dynamic content
   aliases: [],

src/commands/stickers/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const stickers = {
   type: 'local',
   name: 'stickers',
-  description: 'Order OpenClaude stickers',
+  description: 'Order Claude Code stickers',
   supportsNonInteractive: false,
   load: () => import('./stickers.js'),
 } satisfies Command

src/commands/thinkback/index.ts

@@ -4,7 +4,7 @@ import { checkStatsigFeatureGate_CACHED_MAY_BE_STALE } from '../../services/anal
 const thinkback = {
   type: 'local-jsx',
   name: 'think-back',
-  description: 'Your 2025 OpenClaude Year in Review',
+  description: 'Your 2025 Claude Code Year in Review',
   isEnabled: () =>
     checkStatsigFeatureGate_CACHED_MAY_BE_STALE('tengu_thinkback'),
   load: () => import('./thinkback.js'),

src/commands/ultraplan.tsx

@@ -115,7 +115,7 @@ function startDetachedPoll(taskId: string, sessionId: string, url: string, getAp
           ultraplanSessionUrl: undefined
         } : prev);
         enqueuePendingNotification({
-          value: [`Ultraplan approved — executing in OpenClaude on the web. Follow along at: ${url}`, '', 'Results will land as a pull request when the remote session finishes. There is nothing to do here.'].join('\n'),
+          value: [`Ultraplan approved — executing in Claude Code on the web. Follow along at: ${url}`, '', 'Results will land as a pull request when the remote session finishes. There is nothing to do here.'].join('\n'),
           mode: 'task-notification'
         });
       } else {
@@ -184,10 +184,10 @@ function startDetachedPoll(taskId: string, sessionId: string, url: string, getAp
 // multi-second teleportToRemote round-trip.
 function buildLaunchMessage(disconnectedBridge?: boolean): string {
   const prefix = disconnectedBridge ? `${REMOTE_CONTROL_DISCONNECTED_MSG} ` : '';
-  return `${DIAMOND_OPEN} ultraplan\n${prefix}Starting OpenClaude on the web…`;
+  return `${DIAMOND_OPEN} ultraplan\n${prefix}Starting Claude Code on the web…`;
 }
 function buildSessionReadyMessage(url: string): string {
-  return `${DIAMOND_OPEN} ultraplan · Monitor progress in OpenClaude on the web ${url}\nYou can continue working — when the ${DIAMOND_OPEN} fills, press ↓ to view results`;
+  return `${DIAMOND_OPEN} ultraplan · Monitor progress in Claude Code on the web ${url}\nYou can continue working — when the ${DIAMOND_OPEN} fills, press ↓ to view results`;
 }
 function buildAlreadyActiveMessage(url: string | undefined): string {
   return url ? `ultraplan: already polling. Open ${url} to check status, or wait for the plan to land here.` : 'ultraplan: already launching. Please wait for the session to start.';
@@ -272,7 +272,7 @@ export async function launchUltraplan(opts: {
     return [
     // Rendered via <Markdown>; raw <message> is tokenized as HTML
     // and dropped. Backslash-escape the brackets.
-    'Usage: /ultraplan \\<prompt\\>, or include "ultraplan" anywhere', 'in your prompt', '', 'Advanced multi-agent plan mode with our most powerful model', '(Opus). Runs in OpenClaude on the web. When the plan is ready,', 'you can execute it in the web session or send it back here.', 'Terminal stays free while the remote plans.', 'Requires /login.', '', `Terms: ${CCR_TERMS_URL}`].join('\n');
+    'Usage: /ultraplan \\<prompt\\>, or include "ultraplan" anywhere', 'in your prompt', '', 'Advanced multi-agent plan mode with our most powerful model', '(Opus). Runs in Claude Code on the web. When the plan is ready,', 'you can execute it in the web session or send it back here.', 'Terminal stays free while the remote plans.', 'Requires /login.', '', `Terms: ${CCR_TERMS_URL}`].join('\n');
   }
 
   // Set synchronously before the detached flow to prevent duplicate launches
@@ -461,7 +461,7 @@ const call: LocalJSXCommandCall = async (onDone, context, args) => {
 export default {
   type: 'local-jsx',
   name: 'ultraplan',
-  description: `~10–30 min · OpenClaude on the web drafts an advanced plan you can edit and approve. See ${CCR_TERMS_URL}`,
+  description: `~10–30 min · Claude Code on the web drafts an advanced plan you can edit and approve. See ${CCR_TERMS_URL}`,
   argumentHint: '<prompt>',
   isEnabled: () => "external" === 'ant',
   load: () => Promise.resolve({

src/commands/usage/index.ts

@@ -4,5 +4,6 @@ export default {
   type: 'local-jsx',
   name: 'usage',
   description: 'Show plan usage limits',
+  availability: ['claude-ai'],
   load: () => import('./usage.js'),
 } satisfies Command

src/components/ClaudeInChromeOnboarding.tsx

@@ -56,7 +56,7 @@ export function ClaudeInChromeOnboarding(t0) {
   }
   let t5;
   if ($[6] !== t4) {
-    t5 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from OpenClaude. You can navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.{t4}</Text>;
+    t5 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from Claude Code. You can navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.{t4}</Text>;
     $[6] = t4;
     $[7] = t5;
   } else {

src/components/ConsoleOAuthFlow.tsx

@@ -262,7 +262,7 @@ export function ConsoleOAuthFlow({
           state: 'success'
         });
         void sendNotification({
-          message: 'OpenClaude login successful',
+          message: 'Claude Code login successful',
           notificationType: 'auth_success'
         }, terminal);
       }
@@ -384,7 +384,7 @@ function OAuthStatusMessage({
     case 'idle': {
       const promptText =
         startingMessage ||
-        'OpenClaude can be used with your Claude subscription or billed based on API usage through your Console account.'
+        'Claude Code can be used with your Claude subscription or billed based on API usage through your Console account.'
 
       const loginOptions = [
         {
@@ -512,7 +512,7 @@ function OAuthStatusMessage({
         <Box flexDirection="column" gap={1}>
           <Box>
             <Spinner />
-            <Text>Creating API key for OpenClaude…</Text>
+            <Text>Creating API key for Claude Code…</Text>
           </Box>
         </Box>
       )

src/components/DesktopUpsell/DesktopUpsellStartup.tsx

@@ -90,7 +90,7 @@ export function DesktopUpsellStartup(t0) {
   let t3;
   if ($[5] === Symbol.for("react.memo_cache_sentinel")) {
     t3 = {
-      label: "Open in Claude desktop app",
+      label: "Open in Claude Code Desktop",
       value: "try" as const
     };
     $[5] = t3;
@@ -120,7 +120,7 @@ export function DesktopUpsellStartup(t0) {
   const options = t5;
   let t6;
   if ($[8] === Symbol.for("react.memo_cache_sentinel")) {
-    t6 = <Box marginBottom={1}><Text>Use OpenClaude in the Claude desktop app for visual diffs, live app preview, parallel sessions, and more.</Text></Box>;
+    t6 = <Box marginBottom={1}><Text>Same Claude Code with visual diffs, live app preview, parallel sessions, and more.</Text></Box>;
     $[8] = t6;
   } else {
     t6 = $[8];
@@ -135,7 +135,7 @@ export function DesktopUpsellStartup(t0) {
   }
   let t8;
   if ($[11] !== handleSelect || $[12] !== t7) {
-    t8 = <PermissionDialog title="Try the Claude desktop app"><Box flexDirection="column" paddingX={2} paddingY={1}>{t6}<Select options={options} onChange={handleSelect} onCancel={t7} /></Box></PermissionDialog>;
+    t8 = <PermissionDialog title="Try Claude Code Desktop"><Box flexDirection="column" paddingX={2} paddingY={1}>{t6}<Select options={options} onChange={handleSelect} onCancel={t7} /></Box></PermissionDialog>;
     $[11] = handleSelect;
     $[12] = t7;
     $[13] = t8;

src/components/HelpV2/HelpV2.tsx

@@ -138,7 +138,7 @@ export function HelpV2(t0) {
   const t5 = insideModal ? undefined : maxHeight;
   let t6;
   if ($[31] !== tabs) {
-    t6 = <Tabs title={false ? "/help" : `OpenClaude v${MACRO.VERSION}`} color="professionalBlue" defaultTab="general">{tabs}</Tabs>;
+    t6 = <Tabs title={false ? "/help" : `Claude Code v${MACRO.VERSION}`} color="professionalBlue" defaultTab="general">{tabs}</Tabs>;
     $[31] = tabs;
     $[32] = t6;
   } else {
@@ -146,7 +146,7 @@ export function HelpV2(t0) {
   }
   let t7;
   if ($[33] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Box marginTop={1}><Text>For more help:{" "}<Link url="https://github.com/Gitlawb/openclaude" /></Text></Box>;
+    t7 = <Box marginTop={1}><Text>For more help:{" "}<Link url="https://code.claude.com/docs/en/overview" /></Text></Box>;
     $[33] = t7;
   } else {
     t7 = $[33];

src/components/IdeOnboardingDialog.tsx

@@ -70,7 +70,7 @@ export function IdeOnboardingDialog(t0) {
   }
   let t6;
   if ($[8] !== ideName) {
-    t6 = <>{t5}<Text>Welcome to OpenClaude for {ideName}</Text></>;
+    t6 = <>{t5}<Text>Welcome to Claude Code for {ideName}</Text></>;
     $[8] = ideName;
     $[9] = t6;
   } else {

src/components/LogoV2/ChannelsNotice.tsx

@@ -135,7 +135,7 @@ export function ChannelsNotice() {
   }
   let t2;
   if ($[24] !== flag) {
-    t2 = <Text dimColor={true}>Experimental · inbound messages will be pushed into this session, this carries prompt injection risks. Restart OpenClaude without {flag} to disable.</Text>;
+    t2 = <Text dimColor={true}>Experimental · inbound messages will be pushed into this session, this carries prompt injection risks. Restart Claude Code without {flag} to disable.</Text>;
     $[24] = flag;
     $[25] = t2;
   } else {

src/components/LogoV2/LogoV2.tsx

@@ -250,8 +250,8 @@ export function LogoV2() {
   }
   const layoutMode = getLayoutMode(columns);
   const userTheme = resolveThemeSetting(getGlobalConfig().theme);
-  const borderTitle = ` ${color("text", userTheme)("OpenClaude")} ${color("inactive", userTheme)(`v${version}`)} `;
-  const compactBorderTitle = color("text", userTheme)(" OpenClaude ");
+  const borderTitle = ` ${color("text", userTheme)("Open Claude")} ${color("inactive", userTheme)(`v${version}`)} `;
+  const compactBorderTitle = color("text", userTheme)(" Open Claude ");
   if (layoutMode === "compact") {
     let welcomeMessage = formatWelcomeMessage(username);
     if (stringWidth(welcomeMessage) > columns - 4) {

src/components/LogoV2/WelcomeV2.tsx

@@ -9,7 +9,7 @@ export function WelcomeV2() {
   if (env.terminal === "Apple_Terminal") {
     let t0;
     if ($[0] !== theme) {
-      t0 = <AppleTerminalWelcomeV2 theme={theme} welcomeMessage="Welcome to OpenClaude" />;
+      t0 = <AppleTerminalWelcomeV2 theme={theme} welcomeMessage="Welcome to Claude Code" />;
       $[0] = theme;
       $[1] = t0;
     } else {
@@ -28,7 +28,7 @@ export function WelcomeV2() {
     let t7;
     let t8;
     if ($[2] === Symbol.for("react.memo_cache_sentinel")) {
-      t0 = <Text><Text color="claude">{"Welcome to OpenClaude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
+      t0 = <Text><Text color="claude">{"Welcome to Open Claude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
       t1 = <Text>{"\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026"}</Text>;
       t2 = <Text>{"                                                          "}</Text>;
       t3 = <Text>{"                                                          "}</Text>;
@@ -113,7 +113,7 @@ export function WelcomeV2() {
   let t5;
   let t6;
   if ($[18] === Symbol.for("react.memo_cache_sentinel")) {
-    t0 = <Text><Text color="claude">{"Welcome to OpenClaude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
+    t0 = <Text><Text color="claude">{"Welcome to Open Claude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
     t1 = <Text>{"\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026"}</Text>;
     t2 = <Text>{"                                                          "}</Text>;
     t3 = <Text>{"     *                                       \u2588\u2588\u2588\u2588\u2588\u2593\u2593\u2591     "}</Text>;

src/components/LogoV2/feedConfigs.tsx

@@ -41,7 +41,7 @@ export function createWhatsNewFeed(releaseNotes: string[]): FeedConfig {
   });
   const emptyMessage = "external" === 'ant' ? 'Unable to fetch latest claude-cli-internal commits' : 'Check /release-notes for recent updates';
   return {
-    title: "external" === 'ant' ? "OpenClaude Updates [internal-only: Latest CC commits]" : "OpenClaude Updates",
+    title: "external" === 'ant' ? "Open Claude Updates [internal-only: Latest CC commits]" : "Open Claude Updates",
     lines,
     footer: lines.length > 0 ? '/release-notes for more' : undefined,
     emptyMessage
@@ -60,7 +60,7 @@ export function createProjectOnboardingFeed(steps: Step[]): FeedConfig {
       text: `${checkmark}${text}`
     };
   });
-  const warningText = getCwd() === homedir() ? 'Note: You have launched openclaude in your home directory. For the best experience, launch it in a project directory instead.' : undefined;
+  const warningText = getCwd() === homedir() ? 'Note: You have launched claude in your home directory. For the best experience, launch it in a project directory instead.' : undefined;
   if (warningText) {
     lines.push({
       text: warningText
@@ -73,7 +73,7 @@ export function createProjectOnboardingFeed(steps: Step[]): FeedConfig {
 }
 export function createGuestPassesFeed(): FeedConfig {
   const reward = getCachedReferrerReward();
-  const subtitle = reward ? `Share OpenClaude and earn ${formatCreditAmount(reward)} of extra usage` : 'Share OpenClaude with friends';
+  const subtitle = reward ? `Share Open Claude and earn ${formatCreditAmount(reward)} of extra usage` : 'Share Open Claude with friends';
   return {
     title: '3 guest passes',
     lines: [],

src/components/ModelPicker.tsx

@@ -265,7 +265,7 @@ export function ModelPicker(t0) {
   } else {
     t15 = $[41];
   }
-  const t16 = headerText ?? "Switch between Claude models. Applies to this session and future OpenClaude sessions. For other/previous model names, specify with --model.";
+  const t16 = headerText ?? "Switch between Claude models. Applies to this session and future Claude Code sessions. For other/previous model names, specify with --model.";
   let t17;
   if ($[42] !== t16) {
     t17 = <Text dimColor={true}>{t16}</Text>;

src/components/Onboarding.tsx

@@ -146,7 +146,7 @@ export function Onboarding({
     steps.push({
       id: 'terminal-setup',
       component: <Box flexDirection="column" gap={1} paddingLeft={1}>
-          <Text bold>Use OpenClaude&apos;s terminal setup?</Text>
+          <Text bold>Use Claude Code&apos;s terminal setup?</Text>
           <Box flexDirection="column" width={70} gap={1}>
             <Text>
               For the optimal coding experience, enable the recommended settings

src/components/OutputStylePicker.tsx

@@ -80,7 +80,7 @@ export function OutputStylePicker(t0) {
   const t6 = !isStandaloneCommand;
   let t7;
   if ($[5] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Box marginTop={1}><Text dimColor={true}>This changes how OpenClaude communicates with you</Text></Box>;
+    t7 = <Box marginTop={1}><Text dimColor={true}>This changes how Claude Code communicates with you</Text></Box>;
     $[5] = t7;
   } else {
     t7 = $[5];

src/components/PromptInput/PromptInput.tsx

@@ -773,7 +773,7 @@ function PromptInput({
     if (feature('ULTRAPLAN') && ultraplanTriggers.length) {
       addNotification({
         key: 'ultraplan-active',
-        text: 'This prompt will launch an ultraplan session in OpenClaude on the web',
+        text: 'This prompt will launch an ultraplan session in Claude Code on the web',
         priority: 'immediate',
         timeoutMs: 5000
       });

src/components/ProviderManager.test.tsx

@@ -110,7 +110,6 @@ const PRESET_ORDER = [
   'Anthropic',
   'Atomic Chat',
   'Azure OpenAI',
-  'Bankr',
   'Codex OAuth',
   'DeepSeek',
   'Google Gemini',
@@ -118,15 +117,12 @@ const PRESET_ORDER = [
   'LM Studio',
   'MiniMax',
   'Mistral',
-  'Moonshot AI - API',
-  'Moonshot AI - Kimi Code',
+  'Moonshot AI',
   'NVIDIA NIM',
   'Ollama',
   'OpenAI',
   'OpenRouter',
   'Together AI',
-  'xAI',
-  'Z.AI - GLM Coding Plan',
   'Custom',
 ] as const
 
@@ -155,7 +151,6 @@ function createDeferred<T>(): {
 
 function mockProviderProfilesModule(options?: {
   addProviderProfile?: (...args: unknown[]) => unknown
-  getActiveProviderProfile?: () => unknown
   getProviderProfiles?: () => unknown[]
   updateProviderProfile?: (...args: unknown[]) => unknown
   setActiveProviderProfile?: (...args: unknown[]) => unknown
@@ -164,7 +159,7 @@ function mockProviderProfilesModule(options?: {
     addProviderProfile: options?.addProviderProfile ?? (() => null),
     applyActiveProviderProfileFromConfig: () => {},
     deleteProviderProfile: () => ({ removed: false, activeProfileId: null }),
-    getActiveProviderProfile: options?.getActiveProviderProfile ?? (() => null),
+    getActiveProviderProfile: () => null,
     getProviderPresetDefaults: (preset: string) =>
       preset === 'ollama'
         ? {
@@ -194,7 +189,6 @@ function mockProviderManagerDependencies(
     addProviderProfile?: (...args: unknown[]) => unknown
     applySavedProfileToCurrentSession?: (...args: unknown[]) => Promise<string | null>
     clearCodexCredentials?: () => { success: boolean; warning?: string }
-    getActiveProviderProfile?: () => unknown
     getProviderProfiles?: () => unknown[]
     probeOllamaGenerationReadiness?: () => Promise<{
       state: 'ready' | 'unreachable' | 'no_models' | 'generation_failed'
@@ -234,7 +228,6 @@ function mockProviderManagerDependencies(
 ): void {
   mockProviderProfilesModule({
     addProviderProfile: options?.addProviderProfile,
-    getActiveProviderProfile: options?.getActiveProviderProfile,
     getProviderProfiles: options?.getProviderProfiles,
     updateProviderProfile: options?.updateProviderProfile,
     setActiveProviderProfile: options?.setActiveProviderProfile,
@@ -337,10 +330,6 @@ async function mountProviderManager(
   options?: {
     mode?: 'first-run' | 'manage'
     onDone?: (result?: unknown) => void
-    onChangeAppState?: (args: {
-      newState: unknown
-      oldState: unknown
-    }) => void
   },
 ): Promise<{
   stdin: PassThrough
@@ -355,7 +344,7 @@ async function mountProviderManager(
   })
 
   root.render(
-    <AppStateProvider onChangeAppState={options?.onChangeAppState}>
+    <AppStateProvider>
       <KeybindingSetup>
         <ProviderManager
           mode={options?.mode ?? 'manage'}
@@ -917,223 +906,6 @@ test('ProviderManager keeps Codex OAuth as next-startup only when activating the
   await mounted.dispose()
 })
 
-test('ProviderManager activating a multi-model provider sets the session model to the primary model', async () => {
-  delete process.env.CLAUDE_CODE_SIMPLE
-  delete process.env.CLAUDE_CODE_USE_GITHUB
-  delete process.env.GITHUB_TOKEN
-  delete process.env.GH_TOKEN
-
-  const multiModelProfile = {
-    id: 'provider_multi_model',
-    provider: 'openai',
-    name: 'Multi Model Provider',
-    baseUrl: 'https://api.openai.com/v1',
-    model: 'gpt-5.4; gpt-5.4-mini',
-    apiKey: 'sk-test',
-  }
-
-  const setActiveProviderProfile = mock(() => multiModelProfile)
-  const appStateChanges: Array<{ newState: any; oldState: any }> = []
-
-  mockProviderManagerDependencies(
-    () => undefined,
-    async () => undefined,
-    {
-      getProviderProfiles: () => [multiModelProfile],
-      setActiveProviderProfile,
-    },
-  )
-
-  const nonce = `${Date.now()}-${Math.random()}`
-  const { ProviderManager } = await import(`./ProviderManager.js?ts=${nonce}`)
-  const mounted = await mountProviderManager(ProviderManager, {
-    onChangeAppState: args => {
-      appStateChanges.push(args as { newState: any; oldState: any })
-    },
-  })
-
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame =>
-      frame.includes('Provider manager') &&
-      frame.includes('Set active provider'),
-  )
-
-  mounted.stdin.write('j')
-  await Bun.sleep(25)
-  mounted.stdin.write('\r')
-
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame =>
-      frame.includes('Set active provider') &&
-      frame.includes('Multi Model Provider'),
-  )
-
-  await Bun.sleep(25)
-  mounted.stdin.write('\r')
-
-  await waitForCondition(() => setActiveProviderProfile.mock.calls.length > 0)
-  await waitForCondition(() =>
-    appStateChanges.some(
-      ({ newState, oldState }) =>
-        newState.mainLoopModel === 'gpt-5.4' &&
-        oldState.mainLoopModel !== newState.mainLoopModel,
-    ),
-  )
-
-  expect(setActiveProviderProfile).toHaveBeenCalledWith('provider_multi_model')
-  expect(
-    appStateChanges.some(
-      ({ newState }) =>
-        newState.mainLoopModel === 'gpt-5.4' &&
-        newState.mainLoopModelForSession === null,
-    ),
-  ).toBe(true)
-  expect(
-    appStateChanges.some(
-      ({ newState }) => newState.mainLoopModel === 'gpt-5.4; gpt-5.4-mini',
-    ),
-  ).toBe(false)
-
-  await mounted.dispose()
-})
-
-test('ProviderManager editing an active multi-model provider keeps app state on the primary model', async () => {
-  delete process.env.CLAUDE_CODE_SIMPLE
-  delete process.env.CLAUDE_CODE_USE_GITHUB
-  delete process.env.GITHUB_TOKEN
-  delete process.env.GH_TOKEN
-
-  const multiModelProfile = {
-    id: 'provider_multi_model',
-    provider: 'openai',
-    name: 'Multi Model Provider',
-    baseUrl: 'https://api.openai.com/v1',
-    model: 'gpt-5.4; gpt-5.4-mini',
-    apiKey: 'sk-test',
-  }
-
-  const updateProviderProfile = mock(() => multiModelProfile)
-  const appStateChanges: Array<{ newState: any; oldState: any }> = []
-
-  mockProviderManagerDependencies(
-    () => undefined,
-    async () => undefined,
-    {
-      getActiveProviderProfile: () => multiModelProfile,
-      getProviderProfiles: () => [multiModelProfile],
-      updateProviderProfile,
-    },
-  )
-
-  const nonce = `${Date.now()}-${Math.random()}`
-  const { ProviderManager } = await import(`./ProviderManager.js?ts=${nonce}`)
-  const mounted = await mountProviderManager(ProviderManager, {
-    onChangeAppState: args => {
-      appStateChanges.push(args as { newState: any; oldState: any })
-    },
-  })
-
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame =>
-      frame.includes('Provider manager') &&
-      frame.includes('Edit provider'),
-  )
-
-  mounted.stdin.write('j')
-  await Bun.sleep(25)
-  mounted.stdin.write('j')
-  await Bun.sleep(25)
-  mounted.stdin.write('\r')
-
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame =>
-      frame.includes('Edit provider') &&
-      frame.includes('Multi Model Provider'),
-  )
-
-  await Bun.sleep(25)
-  mounted.stdin.write('\r')
-
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame =>
-      frame.includes('Edit provider profile') &&
-      frame.includes('Step 1 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 2 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 3 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 4 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 5 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 6 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-  await waitForFrameOutput(
-    mounted.getOutput,
-    frame => frame.includes('Step 7 of 7'),
-  )
-
-  mounted.stdin.write('\r')
-
-  await waitForCondition(() => updateProviderProfile.mock.calls.length > 0)
-  await waitForCondition(() =>
-    appStateChanges.some(
-      ({ newState, oldState }) =>
-        newState.mainLoopModel === 'gpt-5.4' &&
-        oldState.mainLoopModel !== newState.mainLoopModel,
-    ),
-  )
-
-  expect(updateProviderProfile).toHaveBeenCalledWith(
-    'provider_multi_model',
-    expect.objectContaining({
-      model: 'gpt-5.4; gpt-5.4-mini',
-    }),
-  )
-  expect(
-    appStateChanges.some(
-      ({ newState }) =>
-        newState.mainLoopModel === 'gpt-5.4' &&
-        newState.mainLoopModelForSession === null,
-    ),
-  ).toBe(true)
-  expect(
-    appStateChanges.some(
-      ({ newState }) => newState.mainLoopModel === 'gpt-5.4; gpt-5.4-mini',
-    ),
-  ).toBe(false)
-
-  await mounted.dispose()
-})
-
 test('ProviderManager resolves Codex OAuth state from async storage without sync reads in render flow', async () => {
   delete process.env.CLAUDE_CODE_SIMPLE
   delete process.env.CLAUDE_CODE_USE_GITHUB

src/components/ProviderManager.tsx

@@ -46,7 +46,6 @@ import {
   rankOllamaModels,
   recommendOllamaModel,
 } from '../utils/providerRecommendation.js'
-import { clearStartupProviderOverrides } from '../utils/providerStartupOverrides.js'
 import { redactUrlForDisplay } from '../utils/urlRedaction.js'
 import { updateSettingsForSource } from '../utils/settings/settings.js'
 import {
@@ -58,10 +57,8 @@ import TextInput from './TextInput.js'
 import { useCodexOAuthFlow } from './useCodexOAuthFlow.js'
 
 export type ProviderManagerResult = {
-  action: 'saved' | 'cancelled' | 'activated'
+  action: 'saved' | 'cancelled'
   activeProfileId?: string
-  activeProviderName?: string
-  activeProviderModel?: string
   message?: string
 }
 
@@ -81,14 +78,7 @@ type Screen =
   | 'select-edit'
   | 'select-delete'
 
-type DraftField =
-  | 'name'
-  | 'baseUrl'
-  | 'model'
-  | 'apiKey'
-  | 'apiFormat'
-  | 'authHeader'
-  | 'authHeaderValue'
+type DraftField = 'name' | 'baseUrl' | 'model' | 'apiKey'
 
 type ProviderDraft = Record<DraftField, string>
 
@@ -134,29 +124,8 @@ const FORM_STEPS: Array<{
   {
     key: 'model',
     label: 'Default model',
-    placeholder: 'e.g. llama3.1:8b or glm-4.7; glm-4.7-flash',
-    helpText: 'Model name(s) to use. Separate multiple with ";" or ","; first is default.',
-  },
-  {
-    key: 'apiFormat',
-    label: 'API mode',
-    placeholder: 'chat_completions',
-    helpText: 'Choose the OpenAI-compatible API surface for this provider.',
-    optional: true,
-  },
-  {
-    key: 'authHeader',
-    label: 'Auth header',
-    placeholder: 'e.g. api-key or X-API-Key',
-    helpText: 'Optional. Header name used for a custom provider key.',
-    optional: true,
-  },
-  {
-    key: 'authHeaderValue',
-    label: 'Auth header value',
-    placeholder: 'Leave empty to use the API key value',
-    helpText: 'Optional. Value sent in the custom auth header.',
-    optional: true,
+    placeholder: 'e.g. llama3.1:8b or glm-4.7, glm-4.7-flash',
+    helpText: 'Model name(s) to use. Separate multiple with commas; first is default.',
   },
   {
     key: 'apiKey',
@@ -182,9 +151,6 @@ function toDraft(profile: ProviderProfile): ProviderDraft {
     baseUrl: profile.baseUrl,
     model: profile.model,
     apiKey: profile.apiKey ?? '',
-    apiFormat: profile.apiFormat ?? 'chat_completions',
-    authHeader: profile.authHeader ?? '',
-    authHeaderValue: profile.authHeaderValue ?? '',
   }
 }
 
@@ -195,9 +161,6 @@ function presetToDraft(preset: ProviderPreset): ProviderDraft {
     baseUrl: defaults.baseUrl,
     model: defaults.model,
     apiKey: defaults.apiKey ?? '',
-    apiFormat: 'chat_completions',
-    authHeader: '',
-    authHeaderValue: '',
   }
 }
 
@@ -211,15 +174,7 @@ function profileSummary(profile: ProviderProfile, isActive: boolean): string {
     models.length <= 3
       ? models.join(', ')
       : `${models[0]}, ${models[1]} + ${models.length - 2} more`
-  const modeInfo =
-    profile.provider === 'openai'
-      ? ` · ${profile.apiFormat === 'responses' ? 'responses' : 'chat/completions'}`
-      : ''
-  const authInfo =
-    profile.provider === 'openai' && profile.authHeader
-      ? ` · ${profile.authHeader} auth`
-      : ''
-  return `${providerKind} · ${profile.baseUrl} · ${modelDisplay}${modeInfo}${authInfo} · ${keyInfo}${activeSuffix}`
+  return `${providerKind} · ${profile.baseUrl} · ${modelDisplay} · ${keyInfo}${activeSuffix}`
 }
 
 function getGithubCredentialSourceFromEnv(
@@ -498,18 +453,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     })
   }, [])
 
-  const formSteps = React.useMemo(
-    () =>
-      draftProvider === 'openai'
-        ? FORM_STEPS
-        : FORM_STEPS.filter(step =>
-            step.key !== 'apiFormat' &&
-            step.key !== 'authHeader' &&
-            step.key !== 'authHeaderValue'
-          ),
-    [draftProvider],
-  )
-  const currentStep = formSteps[formStepIndex] ?? formSteps[0] ?? FORM_STEPS[0]
+  const currentStep = FORM_STEPS[formStepIndex] ?? FORM_STEPS[0]
   const currentStepKey = currentStep.key
   const currentValue = draft[currentStepKey]
 
@@ -727,7 +671,17 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
   }
 
   function clearStartupProviderOverrideFromUserSettings(): string | null {
-    return clearStartupProviderOverrides()
+    const { error } = updateSettingsForSource('userSettings', {
+      env: {
+        CLAUDE_CODE_USE_OPENAI: undefined as any,
+        CLAUDE_CODE_USE_GEMINI: undefined as any,
+        CLAUDE_CODE_USE_GITHUB: undefined as any,
+        CLAUDE_CODE_USE_BEDROCK: undefined as any,
+        CLAUDE_CODE_USE_VERTEX: undefined as any,
+        CLAUDE_CODE_USE_FOUNDRY: undefined as any,
+      },
+    })
+    return error ? error.message : null
   }
 
   function buildCodexOAuthActivationMessage(options: {
@@ -814,14 +768,12 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
           mainLoopModelForSession: null,
         }))
         refreshProfiles()
+        setAppState(prev => ({
+          ...prev,
+          mainLoopModel: GITHUB_PROVIDER_DEFAULT_MODEL,
+        }))
         setStatusMessage(`Active provider: ${GITHUB_PROVIDER_LABEL}`)
         setIsActivating(false)
-        onDone({
-          action: 'activated',
-          activeProviderName: GITHUB_PROVIDER_LABEL,
-          activeProviderModel: GITHUB_PROVIDER_DEFAULT_MODEL,
-          message: `Provider switched to ${GITHUB_PROVIDER_LABEL} (${GITHUB_PROVIDER_DEFAULT_MODEL})`,
-        })
         returnToMenu()
         return
       }
@@ -837,14 +789,19 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       // Update the session model to the new provider's first model.
       // persistActiveProviderProfileModel (called by onChangeAppState) will
       // not overwrite the multi-model list because it checks if the model
-      // is already in the provider's configured model list.
+      // is already in the profile's comma-separated model list.
       const newModel = getPrimaryModel(active.model)
       setAppState(prev => ({
         ...prev,
         mainLoopModel: newModel,
+      }))
+
+      providerLabel = active.name
+      setAppState(prev => ({
+        ...prev,
+        mainLoopModel: active.model,
         mainLoopModelForSession: null,
       }))
-      providerLabel = active.name
       const settingsOverrideError =
         clearStartupProviderOverrideFromUserSettings()
       const isActiveCodexOAuth = isCodexOAuthProfile(
@@ -856,7 +813,8 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         : null
 
       refreshProfiles()
-      const activationMessage = isActiveCodexOAuth
+      setStatusMessage(
+        isActiveCodexOAuth
           ? buildCodexOAuthActivationMessage({
               prefix: `Active provider: ${active.name}`,
               activationWarning,
@@ -869,16 +827,9 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
             })
           : settingsOverrideError
             ? `Active provider: ${active.name}. Warning: could not clear startup provider override (${settingsOverrideError}).`
-          : `Active provider: ${active.name}`
-      setStatusMessage(activationMessage)
+            : `Active provider: ${active.name}`,
+      )
       setIsActivating(false)
-      onDone({
-        action: 'activated',
-        activeProfileId: active.id,
-        activeProviderName: active.name,
-        activeProviderModel: newModel,
-        message: `Provider switched to ${active.name} (${newModel})`,
-      })
       returnToMenu()
     } catch (error) {
       refreshProfiles()
@@ -993,9 +944,6 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       baseUrl: defaults.baseUrl,
       model: defaults.model,
       apiKey: defaults.apiKey ?? '',
-      apiFormat: 'chat_completions',
-      authHeader: '',
-      authHeaderValue: '',
     }
     setEditingProfileId(null)
     setDraftProvider(defaults.provider ?? 'openai')
@@ -1042,22 +990,6 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       baseUrl: nextDraft.baseUrl,
       model: nextDraft.model,
       apiKey: nextDraft.apiKey,
-      apiFormat:
-        draftProvider === 'openai' && nextDraft.apiFormat === 'responses'
-          ? 'responses'
-          : 'chat_completions',
-      authHeader:
-        draftProvider === 'openai' && nextDraft.authHeader
-          ? nextDraft.authHeader
-          : undefined,
-      authScheme:
-        draftProvider === 'openai' && nextDraft.authHeader
-          ? (nextDraft.authHeader.toLowerCase() === 'authorization' ? 'bearer' : 'raw')
-          : undefined,
-      authHeaderValue:
-        draftProvider === 'openai' && nextDraft.authHeaderValue
-          ? nextDraft.authHeaderValue
-          : undefined,
     }
 
     const saved = editingProfileId
@@ -1073,7 +1005,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     if (isActiveSavedProfile) {
       setAppState(prev => ({
         ...prev,
-        mainLoopModel: getPrimaryModel(saved.model),
+        mainLoopModel: saved.model,
         mainLoopModelForSession: null,
       }))
     }
@@ -1280,9 +1212,9 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     setDraft(nextDraft)
     setErrorMessage(undefined)
 
-    if (formStepIndex < formSteps.length - 1) {
+    if (formStepIndex < FORM_STEPS.length - 1) {
       const nextIndex = formStepIndex + 1
-      const nextKey = formSteps[nextIndex]?.key ?? 'name'
+      const nextKey = FORM_STEPS[nextIndex]?.key ?? 'name'
       setFormStepIndex(nextIndex)
       setCursorOffset(nextDraft[nextKey].length)
       return
@@ -1296,7 +1228,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
 
     if (formStepIndex > 0) {
       const nextIndex = formStepIndex - 1
-      const nextKey = formSteps[nextIndex]?.key ?? 'name'
+      const nextKey = FORM_STEPS[nextIndex]?.key ?? 'name'
       setFormStepIndex(nextIndex)
       setCursorOffset(draft[nextKey].length)
       return
@@ -1347,11 +1279,6 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         label: 'Azure OpenAI',
         description: 'Azure OpenAI endpoint (model=deployment name)',
       },
-      {
-        value: 'bankr',
-        label: 'Bankr',
-        description: 'Bankr LLM Gateway (OpenAI-compatible)',
-      },
       ...(canUseCodexOAuth
         ? [
             {
@@ -1394,13 +1321,8 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       },
       {
         value: 'moonshotai',
-        label: 'Moonshot AI - API',
-        description: 'Moonshot AI - API endpoint',
-      },
-      {
-        value: 'kimi-code',
-        label: 'Moonshot AI - Kimi Code',
-        description: 'Moonshot AI - Kimi Code Subscription endpoint',
+        label: 'Moonshot AI',
+        description: 'Kimi OpenAI-compatible endpoint',
       },
       {
         value: 'nvidia-nim',
@@ -1427,16 +1349,6 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         label: 'Together AI',
         description: 'Together chat/completions endpoint',
       },
-      {
-        value: 'xai',
-        label: 'xAI',
-        description: 'xAI Grok OpenAI-compatible endpoint',
-      },
-      {
-        value: 'zai',
-        label: 'Z.AI - GLM Coding Plan',
-        description: 'Z.AI GLM coding subscription endpoint',
-      },
       {
         value: 'custom',
         label: 'Custom',
@@ -1501,33 +1413,8 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
             : 'OpenAI-compatible API'}
         </Text>
         <Text dimColor>
-          Step {formStepIndex + 1} of {formSteps.length}: {currentStep.label}
+          Step {formStepIndex + 1} of {FORM_STEPS.length}: {currentStep.label}
         </Text>
-        {currentStepKey === 'apiFormat' ? (
-          <Select
-            options={[
-              {
-                value: 'chat_completions',
-                label: 'Chat Completions',
-                description: 'Use /chat/completions for broad OpenAI-compatible support',
-              },
-              {
-                value: 'responses',
-                label: 'Responses',
-                description: 'Use /responses for providers that support the Responses API',
-              },
-            ]}
-            defaultValue={
-              currentValue === 'responses' ? 'responses' : 'chat_completions'
-            }
-            defaultFocusValue={
-              currentValue === 'responses' ? 'responses' : 'chat_completions'
-            }
-            onChange={value => handleFormSubmit(value)}
-            onCancel={handleBackFromForm}
-            visibleOptionCount={2}
-          />
-        ) : (
         <Box flexDirection="row" gap={1}>
           <Text>{figures.pointer}</Text>
           <TextInput
@@ -1542,18 +1429,12 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
             focus={true}
             showCursor={true}
             placeholder={`${currentStep.placeholder}${figures.ellipsis}`}
-              mask={
-                currentStepKey === 'apiKey' ||
-                currentStepKey === 'authHeaderValue'
-                  ? '*'
-                  : undefined
-              }
+            mask={currentStepKey === 'apiKey' ? '*' : undefined}
             columns={80}
             cursorOffset={cursorOffset}
             onChangeCursorOffset={setCursorOffset}
           />
         </Box>
-        )}
         {errorMessage && <Text color="error">{errorMessage}</Text>}
         <Text dimColor>
           Press Enter to continue. Press Esc to go back.

src/components/ResumeTask.tsx

@@ -119,17 +119,17 @@ export function ResumeTask({
     return <Box flexDirection="column" padding={1}>
         <Box flexDirection="row">
           <Spinner />
-          <Text bold>Loading OpenClaude sessions…</Text>
+          <Text bold>Loading Claude Code sessions…</Text>
         </Box>
         <Text dimColor>
-          {retrying ? 'Retrying…' : 'Fetching your OpenClaude sessions…'}
+          {retrying ? 'Retrying…' : 'Fetching your Claude Code sessions…'}
         </Text>
       </Box>;
   }
   if (loadErrorType) {
     return <Box flexDirection="column" padding={1}>
         <Text bold color="error">
-          Error loading OpenClaude sessions
+          Error loading Claude Code sessions
         </Text>
 
         {renderErrorSpecificGuidance(loadErrorType)}
@@ -143,7 +143,7 @@ export function ResumeTask({
   if (sessions.length === 0) {
     return <Box flexDirection="column" padding={1}>
         <Text bold>
-          No OpenClaude sessions found
+          No Claude Code sessions found
           {currentRepo && <Text> for {currentRepo}</Text>}
         </Text>
         <Box marginTop={1}>
@@ -261,7 +261,7 @@ function renderErrorSpecificGuidance(errorType: LoadErrorType): React.ReactNode
         </Box>;
     case 'other':
       return <Box marginY={1} flexDirection="row">
-          <Text dimColor>Sorry, OpenClaude encountered an error</Text>
+          <Text dimColor>Sorry, Claude Code encountered an error</Text>
         </Box>;
   }
 }

src/components/Settings/Config.tsx

@@ -299,26 +299,6 @@ export function Config({
         enabled: toolHistoryCompressionEnabled
       });
     }
-  }, {
-    id: 'showCacheStats',
-    label: 'Cache stats display',
-    value: globalConfig.showCacheStats,
-    options: ['off', 'compact', 'full'],
-    type: 'enum' as const,
-    onChange(mode: string) {
-      const showCacheStats = (mode === 'off' || mode === 'compact' || mode === 'full' ? mode : 'compact') as 'off' | 'compact' | 'full';
-      saveGlobalConfig(current_cs => ({
-        ...current_cs,
-        showCacheStats
-      }));
-      setGlobalConfig({
-        ...getGlobalConfig(),
-        showCacheStats
-      });
-      logEvent('tengu_show_cache_stats_setting_changed', {
-        mode: showCacheStats as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS
-      });
-    }
   }, {
     id: 'spinnerTipsEnabled',
     label: 'Show tips',

src/components/Settings/MiniMaxUsage.tsx

@@ -1,249 +0,0 @@
-import * as React from 'react'
-import { useEffect, useState } from 'react'
-
-import { useTerminalSize } from '../../hooks/useTerminalSize.js'
-import { Box, Text } from '../../ink.js'
-import { useKeybinding } from '../../keybindings/useKeybinding.js'
-import {
-  buildMiniMaxUsageRows,
-  fetchMiniMaxUsage,
-  type MiniMaxUsageData,
-  type MiniMaxUsageRow,
-} from '../../services/api/minimaxUsage.js'
-import { logError } from '../../utils/log.js'
-import { ConfigurableShortcutHint } from '../ConfigurableShortcutHint.js'
-import { Byline } from '../design-system/Byline.js'
-import { ProgressBar } from '../design-system/ProgressBar.js'
-
-const RESET_COUNTDOWN_REFRESH_MS = 30_000
-const PROGRESS_BAR_WIDTH = 18
-
-type MiniMaxUsageLimitBarProps = {
-  label: string
-  usedPercent: number
-  resetsAt?: string
-  extraSubtext?: string
-  maxWidth: number
-  nowMs: number
-}
-
-function formatCountdownDuration(ms: number): string {
-  const totalMinutes = Math.max(1, Math.ceil(ms / 60_000))
-  const days = Math.floor(totalMinutes / 1_440)
-  const hours = Math.floor((totalMinutes % 1_440) / 60)
-  const minutes = totalMinutes % 60
-
-  if (days > 0) {
-    return hours > 0 ? `${days}d ${hours}h` : `${days}d`
-  }
-
-  if (hours > 0) {
-    return minutes > 0 ? `${hours}h ${minutes}m` : `${hours}h`
-  }
-
-  return `${minutes}m`
-}
-
-function formatResetCountdown(
-  resetsAt: string | undefined,
-  nowMs: number,
-): string | undefined {
-  if (!resetsAt) return undefined
-
-  const resetMs = Date.parse(resetsAt)
-  if (!Number.isFinite(resetMs)) return undefined
-
-  const remainingMs = resetMs - nowMs
-  if (remainingMs <= 0) {
-    return 'Resetting now'
-  }
-
-  return `Resets in ${formatCountdownDuration(remainingMs)}`
-}
-
-function MiniMaxUsageLimitBar({
-  label,
-  usedPercent,
-  resetsAt,
-  extraSubtext,
-  maxWidth,
-  nowMs,
-}: MiniMaxUsageLimitBarProps): React.ReactNode {
-  const normalizedUsedPercent = Math.max(0, Math.min(100, usedPercent))
-  const usedText = `${Math.floor(normalizedUsedPercent)}% used`
-  const resetText = formatResetCountdown(resetsAt, nowMs)
-  const details = [usedText, extraSubtext].filter(
-    (part): part is string => Boolean(part),
-  )
-
-  return (
-    <Box flexDirection="column">
-      <Text>
-        <Text bold>{label}</Text>
-        {resetText ? <Text dimColor> · {resetText}</Text> : null}
-      </Text>
-      <Box flexDirection="row" gap={1}>
-        <ProgressBar
-          ratio={normalizedUsedPercent / 100}
-          width={Math.min(PROGRESS_BAR_WIDTH, Math.max(1, maxWidth))}
-          fillColor="rate_limit_fill"
-          emptyColor="rate_limit_empty"
-        />
-        {details.length > 0 ? <Text dimColor>{details.join(' · ')}</Text> : null}
-      </Box>
-    </Box>
-  )
-}
-
-function MiniMaxUsageTextRow({
-  label,
-  value,
-}: Extract<MiniMaxUsageRow, { kind: 'text' }>): React.ReactNode {
-  if (!value) {
-    return <Text bold>{label}</Text>
-  }
-
-  return (
-    <Text>
-      <Text bold>{label}</Text>
-      <Text dimColor> · {value}</Text>
-    </Text>
-  )
-}
-
-export function MiniMaxUsage(): React.ReactNode {
-  const [usage, setUsage] = useState<MiniMaxUsageData | null>(null)
-  const [error, setError] = useState<string | null>(null)
-  const [isLoading, setIsLoading] = useState(true)
-  const [nowMs, setNowMs] = useState(() => Date.now())
-  const { columns } = useTerminalSize()
-  const availableWidth = columns - 2
-  const maxWidth = Math.min(availableWidth, 80)
-
-  const loadUsage = React.useCallback(async () => {
-    setIsLoading(true)
-    setError(null)
-
-    try {
-      setUsage(await fetchMiniMaxUsage())
-    } catch (err) {
-      logError(err as Error)
-      setError(
-        err instanceof Error ? err.message : 'Failed to load MiniMax usage',
-      )
-    } finally {
-      setIsLoading(false)
-    }
-  }, [])
-
-  useEffect(() => {
-    void loadUsage()
-  }, [loadUsage])
-
-  useEffect(() => {
-    const interval = setInterval(() => {
-      setNowMs(Date.now())
-    }, RESET_COUNTDOWN_REFRESH_MS)
-
-    return () => clearInterval(interval)
-  }, [])
-
-  useKeybinding(
-    'settings:retry',
-    () => {
-      void loadUsage()
-    },
-    {
-      context: 'Settings',
-      isActive: !!error && !isLoading,
-    },
-  )
-
-  if (error) {
-    return (
-      <Box flexDirection="column" gap={1}>
-        <Text color="error">Error: {error}</Text>
-        <Text dimColor>
-          <Byline>
-            <ConfigurableShortcutHint
-              action="settings:retry"
-              context="Settings"
-              fallback="r"
-              description="retry"
-            />
-            <ConfigurableShortcutHint
-              action="confirm:no"
-              context="Settings"
-              fallback="Esc"
-              description="cancel"
-            />
-          </Byline>
-        </Text>
-      </Box>
-    )
-  }
-
-  if (!usage) {
-    return (
-      <Box flexDirection="column" gap={1}>
-        <Text dimColor>Loading MiniMax usage data…</Text>
-        <Text dimColor>
-          <ConfigurableShortcutHint
-            action="confirm:no"
-            context="Settings"
-            fallback="Esc"
-            description="cancel"
-          />
-        </Text>
-      </Box>
-    )
-  }
-
-  const rows =
-    usage.availability === 'available'
-      ? buildMiniMaxUsageRows(usage.snapshots)
-      : []
-
-  return (
-    <Box flexDirection="column" gap={1} width="100%">
-      {usage.planType ? <Text dimColor>Plan: {usage.planType}</Text> : null}
-
-      {usage.availability === 'unknown' ? (
-        <Text dimColor>{usage.message}</Text>
-      ) : rows.length === 0 ? (
-        <Text dimColor>
-          No MiniMax usage windows were returned for this account.
-        </Text>
-      ) : null}
-
-      {rows.map((row, index) =>
-        row.kind === 'window' ? (
-          <MiniMaxUsageLimitBar
-            key={`${row.label}-${index}`}
-            label={row.label}
-            usedPercent={row.usedPercent}
-            resetsAt={row.resetsAt}
-            extraSubtext={row.extraSubtext}
-            maxWidth={maxWidth}
-            nowMs={nowMs}
-          />
-        ) : (
-          <MiniMaxUsageTextRow
-            key={`${row.label}-${index}`}
-            label={row.label}
-            value={row.value}
-          />
-        ),
-      )}
-
-      <Text dimColor>
-        <ConfigurableShortcutHint
-          action="confirm:no"
-          context="Settings"
-          fallback="Esc"
-          description="cancel"
-        />
-      </Text>
-    </Box>
-  )
-}

src/components/Settings/UnsupportedUsage.tsx

@@ -1,28 +0,0 @@
-import * as React from 'react'
-
-import { Box, Text } from '../../ink.js'
-import { ConfigurableShortcutHint } from '../ConfigurableShortcutHint.js'
-
-type UnsupportedUsageProps = {
-  providerLabel: string
-}
-
-export function UnsupportedUsage({
-  providerLabel,
-}: UnsupportedUsageProps): React.ReactNode {
-  return (
-    <Box flexDirection="column" gap={1}>
-      <Text dimColor>
-        Usage details are not currently available for {providerLabel}.
-      </Text>
-      <Text dimColor>
-        <ConfigurableShortcutHint
-          action="confirm:no"
-          context="Settings"
-          fallback="Esc"
-          description="cancel"
-        />
-      </Text>
-    </Box>
-  )
-}

src/components/Settings/Usage.tsx

@@ -17,8 +17,6 @@ import { Byline } from '../design-system/Byline.js';
 import { ProgressBar } from '../design-system/ProgressBar.js';
 import { isEligibleForOverageCreditGrant, OverageCreditUpsell } from '../LogoV2/OverageCreditUpsell.js';
 import { CodexUsage } from './CodexUsage.js';
-import { MiniMaxUsage } from './MiniMaxUsage.js';
-import { UnsupportedUsage } from './UnsupportedUsage.js';
 type LimitBarProps = {
   title: string;
   limit: RateLimit;
@@ -268,26 +266,9 @@ function AnthropicUsage(): React.ReactNode {
     </Box>;
 }
 export function Usage(): React.ReactNode {
-  const provider = getAPIProvider();
-  if (provider === 'codex') {
+  if (getAPIProvider() === 'codex') {
     return <CodexUsage />;
   }
-  if (provider === 'minimax') {
-    return <MiniMaxUsage />;
-  }
-  if (provider !== 'firstParty') {
-    const providerLabel = {
-      openai: 'this OpenAI-compatible provider',
-      gemini: 'Google Gemini',
-      github: 'GitHub Models',
-      mistral: 'Mistral',
-      'nvidia-nim': 'NVIDIA NIM',
-      bedrock: 'AWS Bedrock',
-      vertex: 'Google Vertex AI',
-      foundry: 'Microsoft Foundry'
-    }[provider] ?? 'this provider';
-    return <UnsupportedUsage providerLabel={providerLabel} />;
-  }
   return <AnthropicUsage />;
 }
 type ExtraUsageSectionProps = {

src/components/StartupScreen.test.ts

@@ -14,7 +14,6 @@ const ENV_KEYS = [
   'GEMINI_MODEL',
   'MISTRAL_MODEL',
   'ANTHROPIC_MODEL',
-  'CLAUDE_MODEL',
   'NVIDIA_NIM',
   'MINIMAX_API_KEY',
 ]
@@ -102,14 +101,9 @@ describe('detectProvider — direct vendor endpoints', () => {
     expect(detectProvider().name).toBe('DeepSeek')
   })
 
-  test('api.kimi.com labels as Moonshot AI - Kimi Code', () => {
-    setupOpenAIMode('https://api.kimi.com/coding/v1', 'kimi-for-coding')
-    expect(detectProvider().name).toBe('Moonshot AI - Kimi Code')
-  })
-
-  test('api.moonshot.cn labels as Moonshot AI - API', () => {
+  test('api.moonshot.cn labels as Moonshot (Kimi)', () => {
     setupOpenAIMode('https://api.moonshot.cn/v1', 'moonshot-v1-8k')
-    expect(detectProvider().name).toBe('Moonshot AI - API')
+    expect(detectProvider().name).toBe('Moonshot (Kimi)')
   })
 
   test('api.mistral.ai labels as Mistral', () => {
@@ -117,11 +111,6 @@ describe('detectProvider — direct vendor endpoints', () => {
     expect(detectProvider().name).toBe('Mistral')
   })
 
-  test('api.z.ai labels as Z.AI GLM', () => {
-    setupOpenAIMode('https://api.z.ai/api/coding/paas/v4', 'GLM-5.1')
-    expect(detectProvider().name).toBe('Z.AI - GLM')
-  })
-
   test('default OpenAI URL + gpt-4o labels as OpenAI', () => {
     setupOpenAIMode('https://api.openai.com/v1', 'gpt-4o')
     expect(detectProvider().name).toBe('OpenAI')
@@ -136,14 +125,9 @@ describe('detectProvider — rawModel fallback when URL is generic', () => {
     expect(detectProvider().name).toBe('DeepSeek')
   })
 
-  test('custom proxy + kimi-for-coding falls back to Moonshot AI - Kimi Code', () => {
-    setupOpenAIMode('https://my-proxy.internal/v1', 'kimi-for-coding')
-    expect(detectProvider().name).toBe('Moonshot AI - Kimi Code')
-  })
-
-  test('custom proxy + kimi-k2 falls back to Moonshot AI - API', () => {
+  test('custom proxy + kimi-k2 falls back to Moonshot (Kimi)', () => {
     setupOpenAIMode('https://my-proxy.internal/v1', 'kimi-k2-instruct')
-    expect(detectProvider().name).toBe('Moonshot AI - API')
+    expect(detectProvider().name).toBe('Moonshot (Kimi)')
   })
 
   test('custom proxy + llama-3.3 falls back to Meta Llama', () => {
@@ -155,21 +139,6 @@ describe('detectProvider — rawModel fallback when URL is generic', () => {
     setupOpenAIMode('https://my-proxy.internal/v1', 'mistral-large-latest')
     expect(detectProvider().name).toBe('Mistral')
   })
-
-  test('custom proxy + exact uppercase GLM ID falls back to Z.AI GLM', () => {
-    setupOpenAIMode('https://my-proxy.internal/v1', 'GLM-5.1')
-    expect(detectProvider().name).toBe('Z.AI - GLM')
-  })
-
-  test('custom proxy + lowercase glm ID stays generic OpenAI', () => {
-    setupOpenAIMode('https://my-proxy.internal/v1', 'glm-5.1')
-    expect(detectProvider().name).toBe('OpenAI')
-  })
-
-  test('DashScope lowercase glm ID is not mislabeled as Z.AI', () => {
-    setupOpenAIMode('https://dashscope.aliyuncs.com/compatible-mode/v1', 'glm-5.1')
-    expect(detectProvider().name).toBe('OpenAI')
-  })
 })
 
 // --- Explicit env flags win over URL heuristics ---
@@ -187,71 +156,3 @@ describe('detectProvider — explicit dedicated-provider env flags', () => {
     expect(detectProvider().name).toBe('MiniMax')
   })
 })
-
-// --- modelOverride from --model flag ---
-
-describe('detectProvider — modelOverride from --model flag', () => {
-  test('modelOverride overrides default Anthropic model', () => {
-    const result = detectProvider('claude-opus-4-6')
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('opus')
-  })
-
-  test('modelOverride alias is resolved for Anthropic', () => {
-    const result = detectProvider('opus')
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('opus')
-  })
-
-  test('modelOverride takes priority over ANTHROPIC_MODEL env var', () => {
-    process.env.ANTHROPIC_MODEL = 'claude-haiku-4-5-20251001'
-    const result = detectProvider('claude-opus-4-6')
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('opus')
-  })
-
-  test('modelOverride takes priority over CLAUDE_MODEL env var', () => {
-    process.env.CLAUDE_MODEL = 'claude-haiku-4-5-20251001'
-    const result = detectProvider('claude-opus-4-6')
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('opus')
-  })
-
-  test('modelOverride works for OpenAI provider', () => {
-    process.env.CLAUDE_CODE_USE_OPENAI = '1'
-    process.env.OPENAI_API_KEY = 'test-key'
-    process.env.OPENAI_MODEL = 'gpt-4o'
-    const result = detectProvider('gpt-4-turbo')
-    expect(result.model).toContain('gpt-4-turbo')
-  })
-
-  test('modelOverride works for Gemini provider', () => {
-    process.env.CLAUDE_CODE_USE_GEMINI = '1'
-    const result = detectProvider('gemini-2.5-pro')
-    expect(result.model).toBe('gemini-2.5-pro')
-  })
-
-  test('modelOverride works for Mistral provider', () => {
-    process.env.CLAUDE_CODE_USE_MISTRAL = '1'
-    const result = detectProvider('mistral-large-latest')
-    expect(result.model).toBe('mistral-large-latest')
-  })
-
-  test('modelOverride works for GitHub provider', () => {
-    process.env.CLAUDE_CODE_USE_GITHUB = '1'
-    const result = detectProvider('gpt-4o')
-    expect(result.model).toContain('gpt-4o')
-  })
-
-  test('undefined modelOverride preserves default behavior', () => {
-    const result = detectProvider(undefined)
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('sonnet')
-  })
-
-  test('no argument preserves default behavior', () => {
-    const result = detectProvider()
-    expect(result.name).toBe('Anthropic')
-    expect(result.model).toContain('sonnet')
-  })
-})

src/components/StartupScreen.ts

@@ -9,7 +9,6 @@ import { isLocalProviderUrl, resolveProviderRequest } from '../services/api/prov
 import { getLocalOpenAICompatibleProviderLabel } from '../utils/providerDiscovery.js'
 import { getSettings_DEPRECATED } from '../utils/settings/settings.js'
 import { parseUserSpecifiedModel } from '../utils/model/model.js'
-import { containsExactZaiGlmModelId, isZaiBaseUrl } from '../utils/zaiProvider.js'
 
 declare const MACRO: { VERSION: string; DISPLAY_VERSION?: string }
 
@@ -84,33 +83,33 @@ const LOGO_CLAUDE = [
 
 // ─── Provider detection ───────────────────────────────────────────────────────
 
-export function detectProvider(modelOverride?: string): { name: string; model: string; baseUrl: string; isLocal: boolean } {
+export function detectProvider(): { name: string; model: string; baseUrl: string; isLocal: boolean } {
   const useGemini = process.env.CLAUDE_CODE_USE_GEMINI === '1' || process.env.CLAUDE_CODE_USE_GEMINI === 'true'
   const useGithub = process.env.CLAUDE_CODE_USE_GITHUB === '1' || process.env.CLAUDE_CODE_USE_GITHUB === 'true'
   const useOpenAI = process.env.CLAUDE_CODE_USE_OPENAI === '1' || process.env.CLAUDE_CODE_USE_OPENAI === 'true'
   const useMistral = process.env.CLAUDE_CODE_USE_MISTRAL === '1' || process.env.CLAUDE_CODE_USE_MISTRAL === 'true'
 
   if (useGemini) {
-    const model = modelOverride || process.env.GEMINI_MODEL || 'gemini-2.0-flash'
+    const model = process.env.GEMINI_MODEL || 'gemini-2.0-flash'
     const baseUrl = process.env.GEMINI_BASE_URL || 'https://generativelanguage.googleapis.com/v1beta/openai'
     return { name: 'Google Gemini', model, baseUrl, isLocal: false }
   }
 
   if (useMistral) {
-    const model = modelOverride || process.env.MISTRAL_MODEL || 'devstral-latest'
+    const model = process.env.MISTRAL_MODEL || 'devstral-latest'
     const baseUrl = process.env.MISTRAL_BASE_URL || 'https://api.mistral.ai/v1'
     return { name: 'Mistral', model, baseUrl, isLocal: false }
   }
 
   if (useGithub) {
-    const model = modelOverride || process.env.OPENAI_MODEL || 'github:copilot'
+    const model = process.env.OPENAI_MODEL || 'github:copilot'
     const baseUrl =
       process.env.OPENAI_BASE_URL || 'https://api.githubcopilot.com'
     return { name: 'GitHub Copilot', model, baseUrl, isLocal: false }
   }
 
   if (useOpenAI) {
-    const rawModel = modelOverride || process.env.OPENAI_MODEL || 'gpt-4o'
+    const rawModel = process.env.OPENAI_MODEL || 'gpt-4o'
     const resolvedRequest = resolveProviderRequest({
       model: rawModel,
       baseUrl: process.env.OPENAI_BASE_URL,
@@ -135,26 +134,16 @@ export function detectProvider(modelOverride?: string): { name: string; model: s
     else if (/azure/i.test(baseUrl)) name = 'Azure OpenAI'
     else if (/nvidia/i.test(baseUrl)) name = 'NVIDIA NIM'
     else if (/minimax/i.test(baseUrl)) name = 'MiniMax'
-    else if (/api\.kimi\.com/i.test(baseUrl)) name = 'Moonshot AI - Kimi Code'
-    else if (/moonshot/i.test(baseUrl)) name = 'Moonshot AI - API'
+    else if (/moonshot/i.test(baseUrl)) name = 'Moonshot (Kimi)'
     else if (/deepseek/i.test(baseUrl)) name = 'DeepSeek'
-    else if (/x\.ai/i.test(baseUrl)) name = 'xAI'
-    else if (isZaiBaseUrl(baseUrl)) name = 'Z.AI - GLM'
     else if (/mistral/i.test(baseUrl)) name = 'Mistral'
     // rawModel fallback — fires only when base URL is generic/custom.
     else if (/nvidia/i.test(rawModel)) name = 'NVIDIA NIM'
     else if (/minimax/i.test(rawModel)) name = 'MiniMax'
-    else if (/\bkimi-for-coding\b/i.test(rawModel))
-      name = 'Moonshot AI - Kimi Code'
-    else if (/\bkimi-k/i.test(rawModel) || /moonshot/i.test(rawModel))
-      name = 'Moonshot AI - API'
+    else if (/kimi/i.test(rawModel)) name = 'Moonshot (Kimi)'
     else if (/deepseek/i.test(rawModel)) name = 'DeepSeek'
-    else if (/grok/i.test(rawModel)) name = 'xAI'
-    else if (containsExactZaiGlmModelId(rawModel)) name = 'Z.AI - GLM'
     else if (/mistral/i.test(rawModel)) name = 'Mistral'
     else if (/llama/i.test(rawModel)) name = 'Meta Llama'
-    else if (/bankr/i.test(baseUrl)) name = 'Bankr'
-    else if (/bankr/i.test(rawModel)) name = 'Bankr'
     else if (isLocal) name = getLocalOpenAICompatibleProviderLabel(baseUrl)
     
     // Resolve model alias to actual model name + reasoning effort
@@ -168,7 +157,7 @@ export function detectProvider(modelOverride?: string): { name: string; model: s
 
   // Default: Anthropic - check settings.model first, then env vars
   const settings = getSettings_DEPRECATED() || {}
-  const modelSetting = modelOverride || settings.model || process.env.ANTHROPIC_MODEL || process.env.CLAUDE_MODEL || 'claude-sonnet-4-6'
+  const modelSetting = settings.model || process.env.ANTHROPIC_MODEL || process.env.CLAUDE_MODEL || 'claude-sonnet-4-6'
   const resolvedModel = parseUserSpecifiedModel(modelSetting)
   const baseUrl = process.env.ANTHROPIC_BASE_URL ?? 'https://api.anthropic.com'
   const isLocal = isLocalProviderUrl(baseUrl)
@@ -184,11 +173,11 @@ function boxRow(content: string, width: number, rawLen: number): string {
 
 // ─── Main ─────────────────────────────────────────────────────────────────────
 
-export function printStartupScreen(modelOverride?: string): void {
+export function printStartupScreen(): void {
   // Skip in non-interactive / CI / print mode
   if (process.env.CI || !process.stdout.isTTY) return
 
-  const p = detectProvider(modelOverride)
+  const p = detectProvider()
   const W = 62
   const out: string[] = []
 

src/components/Stats.tsx

@@ -94,7 +94,7 @@ export function Stats(t0) {
   const allTimePromise = t1;
   let t2;
   if ($[1] === Symbol.for("react.memo_cache_sentinel")) {
-    t2 = <Box marginTop={1}><Spinner /><Text> Loading your OpenClaude stats…</Text></Box>;
+    t2 = <Box marginTop={1}><Spinner /><Text> Loading your Claude Code stats…</Text></Box>;
     $[1] = t2;
   } else {
     t2 = $[1];
@@ -242,7 +242,7 @@ function StatsContent(t0) {
   if (allTimeResult.type === "empty") {
     let t7;
     if ($[15] === Symbol.for("react.memo_cache_sentinel")) {
-      t7 = <Box marginTop={1}><Text color="warning">No stats available yet. Start using OpenClaude!</Text></Box>;
+      t7 = <Box marginTop={1}><Text color="warning">No stats available yet. Start using Claude Code!</Text></Box>;
       $[15] = t7;
     } else {
       t7 = $[15];

src/components/TeleportRepoMismatchDialog.tsx

@@ -73,7 +73,7 @@ export function TeleportRepoMismatchDialog(t0) {
   const options = t2;
   let t3;
   if ($[8] !== availablePaths.length || $[9] !== errorMessage || $[10] !== handleChange || $[11] !== options || $[12] !== targetRepo || $[13] !== validating) {
-    t3 = availablePaths.length > 0 ? <><Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text>Open OpenClaude in <Text bold={true}>{targetRepo}</Text>:</Text></Box>{validating ? <Box><Spinner /><Text> Validating repository…</Text></Box> : <Select options={options} onChange={value_0 => void handleChange(value_0)} />}</> : <Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text dimColor={true}>Run openclaude --teleport from a checkout of {targetRepo}</Text></Box>;
+    t3 = availablePaths.length > 0 ? <><Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text>Open Claude Code in <Text bold={true}>{targetRepo}</Text>:</Text></Box>{validating ? <Box><Spinner /><Text> Validating repository…</Text></Box> : <Select options={options} onChange={value_0 => void handleChange(value_0)} />}</> : <Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text dimColor={true}>Run claude --teleport from a checkout of {targetRepo}</Text></Box>;
     $[8] = availablePaths.length;
     $[9] = errorMessage;
     $[10] = handleChange;

src/components/TrustDialog/TrustDialog.tsx

@@ -206,7 +206,7 @@ export function TrustDialog(t0) {
   if ($[20] === Symbol.for("react.memo_cache_sentinel")) {
     t16 = <Text bold={true}>{getFsImplementation().cwd()}</Text>;
     t17 = <Text>Quick safety check: Is this a project you created or one you trust? (Like your own code, a well-known open source project, or work from your team). If not, take a moment to review what{"'"}s in this folder first.</Text>;
-    t18 = <Text>OpenClaude{"'"}ll be able to read, edit, and execute files here.</Text>;
+    t18 = <Text>Claude Code{"'"}ll be able to read, edit, and execute files here.</Text>;
     $[20] = t16;
     $[21] = t17;
     $[22] = t18;

src/components/mcp/ElicitationDialog.tsx

@@ -254,7 +254,7 @@ function ElicitationFormDialog({
   // Text fields are always in edit mode when focused — no Enter-to-edit step.
   const isEditingTextField = currentFieldIsText && !focusedButton;
   useRegisterOverlay('elicitation');
-  useNotifyAfterTimeout('OpenClaude needs your input', 'elicitation_dialog');
+  useNotifyAfterTimeout('Claude Code needs your input', 'elicitation_dialog');
 
   // Sync textInputValue when the focused field changes
   const syncTextInput = useCallback((fieldIndex: number | undefined) => {
@@ -1004,7 +1004,7 @@ function ElicitationURLDialog({
   const phaseRef = useRef<'prompt' | 'waiting'>('prompt');
   const [focusedButton, setFocusedButton] = useState<'accept' | 'decline' | 'open' | 'action' | 'cancel'>('accept');
   const showCancel = waitingState?.showCancel ?? false;
-  useNotifyAfterTimeout('OpenClaude needs your input', 'elicitation_url_dialog');
+  useNotifyAfterTimeout('Claude Code needs your input', 'elicitation_url_dialog');
   useRegisterOverlay('elicitation-url');
 
   // Keep refs in sync for use in abort handler (avoids re-registering listener)

src/components/mcp/MCPRemoteServerMenu.tsx

@@ -102,9 +102,9 @@ export function MCPRemoteServerMenu({
       if (success) {
         onComplete?.(`Authentication successful. Connected to ${server.name}.`);
       } else if (result.client.type === 'needs-auth') {
-        onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart OpenClaude.');
+        onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart Claude Code.');
       } else {
-        onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart OpenClaude for the changes to take effect.');
+        onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart Claude Code for the changes to take effect.');
       }
     } catch (err) {
       logEvent('tengu_claudeai_mcp_auth_completed', {
@@ -281,11 +281,11 @@ export function MCPRemoteServerMenu({
           const message = isEffectivelyAuthenticated ? `Authentication successful. Reconnected to ${server.name}.` : `Authentication successful. Connected to ${server.name}.`;
           onComplete?.(message);
         } else if (result_0.client.type === 'needs-auth') {
-          onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart OpenClaude.');
+          onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart Claude Code.');
         } else {
           // result.client.type === 'failed'
           logMCPDebug(server.name, `Reconnection failed after authentication`);
-          onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart OpenClaude for the changes to take effect.');
+          onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart Claude Code for the changes to take effect.');
         }
       }
     } catch (err_1) {

src/components/mcp/MCPSettings.tsx

@@ -147,7 +147,7 @@ export function MCPSettings(t0) {
         return;
       }
       if (servers.length === 0 && agentMcpServers.length === 0) {
-        onComplete("No MCP servers configured. Please run /doctor if this is unexpected. Otherwise, run `openclaude mcp --help` or visit https://github.com/Gitlawb/openclaude to learn more.");
+        onComplete("No MCP servers configured. Please run /doctor if this is unexpected. Otherwise, run `claude mcp --help` or visit https://code.claude.com/docs/en/mcp to learn more.");
       }
     };
     t8 = [servers.length, filteredClients.length, agentMcpServers.length, onComplete];

src/components/permissions/ComputerUseApproval/ComputerUseApproval.tsx

@@ -161,7 +161,7 @@ function ComputerUseTccPanel(t0) {
   }
   let t7;
   if ($[15] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Text dimColor={true}>Grant the missing permissions in System Settings, then select "Try again". macOS may require you to restart OpenClaude after granting Screen Recording.</Text>;
+    t7 = <Text dimColor={true}>Grant the missing permissions in System Settings, then select "Try again". macOS may require you to restart Claude Code after granting Screen Recording.</Text>;
     $[15] = t7;
   } else {
     t7 = $[15];

src/components/permissions/ExitPlanModePermissionRequest/ExitPlanModePermissionRequest.tsx

@@ -730,7 +730,7 @@ export function buildPlanApprovalOptions({
   });
   if (showUltraplan) {
     options.push({
-      label: 'No, refine with Ultraplan on OpenClaude on the web',
+      label: 'No, refine with Ultraplan on Claude Code on the web',
       value: 'ultraplan'
     });
   }

src/components/permissions/PermissionRequest.tsx

@@ -128,18 +128,18 @@ export type ToolUseConfirm<Input extends AnyObject = AnyObject> = {
 function getNotificationMessage(toolUseConfirm: ToolUseConfirm): string {
   const toolName = toolUseConfirm.tool.userFacingName(toolUseConfirm.input as never);
   if (toolUseConfirm.tool === ExitPlanModeV2Tool) {
-    return 'OpenClaude needs your approval for the plan';
+    return 'Claude Code needs your approval for the plan';
   }
   if (toolUseConfirm.tool === EnterPlanModeTool) {
-    return 'OpenClaude wants to enter plan mode';
+    return 'Claude Code wants to enter plan mode';
   }
   if (feature('REVIEW_ARTIFACT') && toolUseConfirm.tool === ReviewArtifactTool) {
-    return 'OpenClaude needs your approval for a review artifact';
+    return 'Claude needs your approval for a review artifact';
   }
   if (!toolName || toolName.trim() === '') {
-    return 'OpenClaude needs your attention';
+    return 'Claude Code needs your attention';
   }
-  return `OpenClaude needs your permission to use ${toolName}`;
+  return `Claude needs your permission to use ${toolName}`;
 }
 
 // TODO: Move this to Tool.renderPermissionRequest

src/components/permissions/rules/AddWorkspaceDirectory.tsx

@@ -40,7 +40,7 @@ function PermissionDescription() {
   const $ = _c(1);
   let t0;
   if ($[0] === Symbol.for("react.memo_cache_sentinel")) {
-    t0 = <Text dimColor={true}>OpenClaude will be able to read files in this directory and make edits when auto-accept edits is on.</Text>;
+    t0 = <Text dimColor={true}>Claude Code will be able to read files in this directory and make edits when auto-accept edits is on.</Text>;
     $[0] = t0;
   } else {
     t0 = $[0];

src/components/permissions/rules/PermissionRuleList.tsx

@@ -388,9 +388,9 @@ function PermissionRulesTab(t0) {
     let t8;
     if ($[10] === Symbol.for("react.memo_cache_sentinel")) {
       t8 = {
-        allow: "OpenClaude won't ask before using allowed tools.",
-        ask: "OpenClaude will always ask for confirmation before using these tools.",
-        deny: "OpenClaude will always reject requests to use denied tools."
+        allow: "Claude Code won't ask before using allowed tools.",
+        ask: "Claude Code will always ask for confirmation before using these tools.",
+        deny: "Claude Code will always reject requests to use denied tools."
       };
       $[10] = t8;
     } else {
@@ -1098,7 +1098,7 @@ export function PermissionRuleList(t0) {
   }
   let t28;
   if ($[89] === Symbol.for("react.memo_cache_sentinel")) {
-    t28 = <Text>OpenClaude can read files in the workspace, and make edits when auto-accept edits is on.</Text>;
+    t28 = <Text>Claude Code can read files in the workspace, and make edits when auto-accept edits is on.</Text>;
     $[89] = t28;
   } else {
     t28 = $[89];

src/components/permissions/rules/RemoveWorkspaceDirectory.tsx

@@ -68,7 +68,7 @@ export function RemoveWorkspaceDirectory(t0) {
   }
   let t4;
   if ($[10] === Symbol.for("react.memo_cache_sentinel")) {
-    t4 = <Text>OpenClaude will no longer have access to files in this directory.</Text>;
+    t4 = <Text>Claude Code will no longer have access to files in this directory.</Text>;
     $[10] = t4;
   } else {
     t4 = $[10];

src/components/tasks/RemoteSessionDetailDialog.tsx

@@ -44,7 +44,7 @@ type Props = {
 export function formatToolUseSummary(name: string, input: unknown): string {
   // plan_ready phase is only reached via ExitPlanMode tool
   if (name === EXIT_PLAN_MODE_V2_TOOL_NAME) {
-    return 'Review the plan in OpenClaude on the web';
+    return 'Review the plan in Claude Code on the web';
   }
   if (!input || typeof input !== 'object') return name;
   // AskUserQuestion: show the question text as a CTA, not the tool name.
@@ -168,7 +168,7 @@ function UltraplanSessionDetail(t0) {
     }
     let t7;
     if ($[12] === Symbol.for("react.memo_cache_sentinel")) {
-      t7 = <Text dimColor={true}>This will terminate the OpenClaude on the web session.</Text>;
+      t7 = <Text dimColor={true}>This will terminate the Claude Code on the web session.</Text>;
       $[12] = t7;
     } else {
       t7 = $[12];
@@ -311,7 +311,7 @@ function UltraplanSessionDetail(t0) {
   let t19;
   if ($[47] === Symbol.for("react.memo_cache_sentinel")) {
     t19 = {
-      label: "Review in OpenClaude on the web",
+      label: "Review in Claude Code on the web",
       value: "open" as const
     };
     $[47] = t19;
@@ -595,13 +595,13 @@ function ReviewSessionDetail(t0) {
   let t3;
   if ($[11] !== completed || $[12] !== onKill || $[13] !== running) {
     t3 = completed ? [{
-      label: "Open in OpenClaude on the web",
+      label: "Open in Claude Code on the web",
       value: "open"
     }, {
       label: "Dismiss",
       value: "dismiss"
     }] : [{
-      label: "Open in OpenClaude on the web",
+      label: "Open in Claude Code on the web",
       value: "open"
     }, ...(onKill && running ? [{
       label: "Stop ultrareview",

src/constants/promptIdentity.test.ts

@@ -11,7 +11,6 @@ import { afterEach, expect, test } from 'bun:test'
   NATIVE_PACKAGE_URL: undefined,
 }
 
-import { clearSystemPromptSections } from './systemPromptSections.js'
 import { getSystemPrompt, DEFAULT_AGENT_PROMPT } from './prompts.js'
 import { CLI_SYSPROMPT_PREFIXES, getCLISyspromptPrefix } from './system.js'
 import { CLAUDE_CODE_GUIDE_AGENT } from '../tools/AgentTool/built-in/claudeCodeGuideAgent.js'
@@ -24,7 +23,6 @@ const originalSimpleEnv = process.env.CLAUDE_CODE_SIMPLE
 
 afterEach(() => {
   process.env.CLAUDE_CODE_SIMPLE = originalSimpleEnv
-  clearSystemPromptSections()
 })
 
 test('CLI identity prefixes describe OpenClaude instead of Claude Code', () => {
@@ -49,21 +47,6 @@ test('simple mode identity describes OpenClaude instead of Claude Code', async (
   expect(prompt[0]).not.toContain("Anthropic's official CLI for Claude")
 })
 
-test('system prompt model identity updates when model changes mid-session', async () => {
-  delete process.env.CLAUDE_CODE_SIMPLE
-  clearSystemPromptSections()
-
-  const firstPrompt = await getSystemPrompt([], 'old-test-model')
-  const secondPrompt = await getSystemPrompt([], 'new-test-model')
-
-  const firstText = firstPrompt.join('\n')
-  const secondText = secondPrompt.join('\n')
-
-  expect(firstText).toContain('You are powered by the model old-test-model.')
-  expect(secondText).toContain('You are powered by the model new-test-model.')
-  expect(secondText).not.toContain('You are powered by the model old-test-model.')
-})
-
 test('built-in agent prompts describe OpenClaude instead of Claude Code', () => {
   expect(DEFAULT_AGENT_PROMPT).toContain('OpenClaude')
   expect(DEFAULT_AGENT_PROMPT).not.toContain('Claude Code')

src/constants/prompts.ts

@@ -496,7 +496,7 @@ ${CYBER_RISK_INSTRUCTION}`,
     systemPromptSection('ant_model_override', () =>
       getAntModelOverrideSection(),
     ),
-    systemPromptSection(`env_info_simple:${model}`, () =>
+    systemPromptSection('env_info_simple', () =>
       computeSimpleEnvInfo(model, additionalWorkingDirectories),
     ),
     systemPromptSection('language', () =>
@@ -519,7 +519,7 @@ ${CYBER_RISK_INSTRUCTION}`,
       'MCP servers connect/disconnect between turns',
     ),
     systemPromptSection('scratchpad', () => getScratchpadInstructions()),
-    systemPromptSection(`frc:${model}`, () => getFunctionResultClearingSection(model)),
+    systemPromptSection('frc', () => getFunctionResultClearingSection(model)),
     systemPromptSection(
       'summarize_tool_results',
       () => SUMMARIZE_TOOL_RESULTS_SECTION,

src/cost-tracker.cacheIntegration.test.ts

@@ -1,128 +0,0 @@
-/**
- * Integration test for cost-tracker → cacheStatsTracker wiring.
- *
- * The unit tests in services/api/cacheMetrics.test.ts and
- * services/api/cacheStatsTracker.test.ts verify that each piece works
- * in isolation. This file verifies that they're ACTUALLY CONNECTED —
- * that `addToTotalSessionCost` resolves the provider, extracts metrics,
- * and records them on the tracker on every call. Without this test, a
- * future refactor could silently unwire the call chain (wrong param
- * order, renamed symbol, removed call) and every individual unit test
- * would still pass while `/cache-stats` showed empty data.
- *
- * We use real state — `resetCostState` + `getCurrentTurnCacheMetrics` —
- * rather than mocking the tracker module. Fewer moving parts, and the
- * test fails for the right reason if anyone breaks the wrapping.
- */
-import { beforeEach, describe, expect, test } from 'bun:test'
-import { addToTotalSessionCost, resetCostState } from './cost-tracker.js'
-import {
-  getCurrentTurnCacheMetrics,
-  getSessionCacheMetrics,
-} from './services/api/cacheStatsTracker.js'
-
-// BetaUsage-compatible shape — minimum fields addToTotalSessionCost
-// needs to run without throwing. Cache fields are the ones we care
-// about here; input/output go into model cost calc.
-function anthropicUsage(partial: {
-  input?: number
-  output?: number
-  cacheRead?: number
-  cacheCreation?: number
-}): Parameters<typeof addToTotalSessionCost>[1] {
-  return {
-    input_tokens: partial.input ?? 0,
-    output_tokens: partial.output ?? 0,
-    cache_read_input_tokens: partial.cacheRead ?? 0,
-    cache_creation_input_tokens: partial.cacheCreation ?? 0,
-    // BetaUsage has several other optional fields; they're not read by
-    // the cache-tracking path so we leave them undefined.
-  } as Parameters<typeof addToTotalSessionCost>[1]
-}
-
-beforeEach(() => {
-  // resetCostState is the wrapped version that ALSO clears the cache
-  // tracker — this line is itself part of what we're verifying.
-  resetCostState()
-})
-
-describe('addToTotalSessionCost → cacheStatsTracker wiring', () => {
-  test('records normalized cache metrics on the tracker for each call', () => {
-    addToTotalSessionCost(
-      0.01,
-      anthropicUsage({
-        input: 200,
-        output: 50,
-        cacheRead: 800,
-        cacheCreation: 100,
-      }),
-      'claude-sonnet-4',
-    )
-
-    const turn = getCurrentTurnCacheMetrics()
-    expect(turn.supported).toBe(true)
-    expect(turn.read).toBe(800)
-    expect(turn.created).toBe(100)
-    // total = fresh(200) + read(800) + created(100) = 1100
-    expect(turn.total).toBe(1_100)
-    // hitRate = read / total = 800 / 1100 ≈ 0.727
-    expect(turn.hitRate).toBeCloseTo(800 / 1_100, 4)
-  })
-
-  test('session aggregate accumulates across multiple API calls', () => {
-    addToTotalSessionCost(
-      0.01,
-      anthropicUsage({ input: 100, cacheRead: 400 }),
-      'claude-sonnet-4',
-    )
-    addToTotalSessionCost(
-      0.02,
-      anthropicUsage({ input: 200, cacheRead: 600 }),
-      'claude-sonnet-4',
-    )
-
-    const session = getSessionCacheMetrics()
-    expect(session.read).toBe(1_000)
-    // total = (100+400) + (200+600) = 1300
-    expect(session.total).toBe(1_300)
-    expect(session.hitRate).toBeCloseTo(1_000 / 1_300, 4)
-  })
-
-  test('cold turn (no cache read/created) still records as supported', () => {
-    addToTotalSessionCost(
-      0.005,
-      anthropicUsage({ input: 500, output: 100 }),
-      'claude-sonnet-4',
-    )
-
-    const turn = getCurrentTurnCacheMetrics()
-    expect(turn.supported).toBe(true)
-    expect(turn.read).toBe(0)
-    expect(turn.created).toBe(0)
-    expect(turn.total).toBe(500)
-    // hitRate computed against a non-zero total is 0, not null — empty
-    // cache on a cacheable provider is a legitimate "no-hit" signal.
-    expect(turn.hitRate).toBe(0)
-  })
-})
-
-describe('resetCostState wrapper also clears cache tracker', () => {
-  test('resetCostState() zeros both cost counters and cache stats', () => {
-    // Populate both systems
-    addToTotalSessionCost(
-      0.01,
-      anthropicUsage({ input: 100, cacheRead: 500 }),
-      'claude-sonnet-4',
-    )
-    expect(getSessionCacheMetrics().read).toBe(500)
-
-    // resetCostState is the WRAPPED version — bootstrap's
-    // resetCostState cleared cost state historically but not cache
-    // stats. The wrapper in cost-tracker.ts adds the second call.
-    resetCostState()
-
-    const session = getSessionCacheMetrics()
-    expect(session.read).toBe(0)
-    expect(session.supported).toBe(false)
-  })
-})

src/cost-tracker.ts

@@ -1,14 +1,5 @@
 import type { BetaUsage as Usage } from '@anthropic-ai/sdk/resources/beta/messages/messages.mjs'
 import chalk from 'chalk'
-import {
-  extractCacheMetrics,
-  resolveCacheProvider,
-} from './services/api/cacheMetrics.js'
-import {
-  recordRequest as recordCacheRequest,
-  resetSessionCacheStats,
-} from './services/api/cacheStatsTracker.js'
-import { getAPIProvider, isGithubNativeAnthropicMode } from './utils/model/providers.js'
 import {
   addToTotalCostState,
   addToTotalLinesChanged,
@@ -31,7 +22,7 @@ import {
   getTotalWebSearchRequests,
   getUsageForModel,
   hasUnknownModelCost,
-  resetCostState as baseResetCostState,
+  resetCostState,
   resetStateForTests,
   setCostStateForRestore,
   setHasUnknownModelCost,
@@ -71,22 +62,12 @@ export {
   formatCost,
   hasUnknownModelCost,
   resetStateForTests,
+  resetCostState,
   setHasUnknownModelCost,
   getModelUsage,
   getUsageForModel,
 }
 
-/**
- * Wraps bootstrap's resetCostState() so /clear, /compact and session
- * switches zero the cache-stats tracker alongside the cost counters.
- * Exported under the same name so existing callers pick up the cache
- * reset without any call-site changes.
- */
-export function resetCostState(): void {
-  baseResetCostState()
-  resetSessionCacheStats()
-}
-
 type StoredCostState = {
   totalCostUSD: number
   totalAPIDuration: number
@@ -270,16 +251,6 @@ function round(number: number, precision: number): number {
   return Math.round(number * precision) / precision
 }
 
-// Env-gated verbose token usage log. Treated as a boolean regardless of
-// value specifics — any truthy-ish string switches it on. `verbose` is the
-// documented keyword but we accept `1`/`true` for ergonomic parity with
-// other OPENCLAUDE_* flags.
-function shouldLogTokenUsageVerbose(): boolean {
-  const v = (process.env.OPENCLAUDE_LOG_TOKEN_USAGE ?? '').trim().toLowerCase()
-  if (!v) return false
-  return v !== '0' && v !== 'false' && v !== 'off'
-}
-
 function addToTotalModelUsage(
   cost: number,
   usage: Usage,
@@ -316,43 +287,6 @@ export function addToTotalSessionCost(
   const modelUsage = addToTotalModelUsage(cost, usage, model)
   addToTotalCostState(cost, modelUsage, model)
 
-  // Record normalized cache metrics for REPL display + /cache-stats.
-  // Resolved from the current process provider — at this point `usage` has
-  // already been Anthropic-shaped by the shim layer, so we feed the
-  // corresponding bucket (anthropic / copilot-claude / openai-like) to the
-  // extractor. For providers that genuinely don't report cache data
-  // (vanilla Copilot, Ollama), resolveCacheProvider steers us to
-  // supported:false so the UI shows "N/A" instead of lying with "0%".
-  const cacheProvider = resolveCacheProvider(getAPIProvider(), {
-    githubNativeAnthropic: isGithubNativeAnthropicMode(model),
-    openAiBaseUrl: process.env.OPENAI_BASE_URL ?? process.env.OPENAI_API_BASE,
-  })
-  const cacheMetrics = extractCacheMetrics(
-    usage as unknown as Record<string, unknown>,
-    cacheProvider,
-  )
-  recordCacheRequest(cacheMetrics, model)
-
-  // Opt-in structured per-request debug log on stderr. Power-user knob, not
-  // shown in the REPL — complements CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT
-  // (which is model-facing). Any truthy value except "0"/"false" enables it.
-  if (shouldLogTokenUsageVerbose()) {
-    process.stderr.write(
-      JSON.stringify({
-        tag: 'openclaude.tokenUsage',
-        model,
-        provider: cacheProvider,
-        input_tokens: usage.input_tokens,
-        output_tokens: usage.output_tokens,
-        cache_read_input_tokens: usage.cache_read_input_tokens ?? 0,
-        cache_creation_input_tokens: usage.cache_creation_input_tokens ?? 0,
-        cache_supported: cacheMetrics.supported,
-        cache_hit_rate: cacheMetrics.hitRate,
-        cost_usd: cost,
-      }) + '\n',
-    )
-  }
-
   const attrs =
     isFastModeEnabled() && usage.speed === 'fast'
       ? { model, speed: 'fast' }

src/entrypoints/cli.tsx

@@ -80,7 +80,7 @@ async function main(): Promise<void> {
   if (args.length === 1 && (args[0] === '--version' || args[0] === '-v' || args[0] === '-V')) {
     // MACRO.VERSION is inlined at build time
     // biome-ignore lint/suspicious/noConsole:: intentional console output
-    console.log(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (OpenClaude)`);
+    console.log(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (Open Claude)`);
     return;
   }
 
@@ -134,13 +134,9 @@ async function main(): Promise<void> {
 
   await validateProviderEnvForStartupOrExit()
 
-  // Parse --model early so the startup screen can display the override
-  const { eagerParseCliFlag } = await import('../utils/cliArgs.js')
-  const earlyModelFlag = eagerParseCliFlag('--model')
-
   // Print the gradient startup screen before the Ink UI loads
   const { printStartupScreen } = await import('../components/StartupScreen.js')
-  printStartupScreen(earlyModelFlag)
+  printStartupScreen()
 
   // For all other paths, load the startup profiler
   const {

src/hooks/notifs/useNpmDeprecationNotification.tsx

@@ -2,7 +2,7 @@ import { isInBundledMode } from 'src/utils/bundledMode.js';
 import { getCurrentInstallationType } from 'src/utils/doctorDiagnostic.js';
 import { isEnvTruthy } from 'src/utils/envUtils.js';
 import { useStartupNotification } from './useStartupNotification.js';
-const NPM_DEPRECATION_MESSAGE = 'OpenClaude has switched from npm to the native installer. Run `openclaude install` or see https://github.com/Gitlawb/openclaude#quick-start for more options.';
+const NPM_DEPRECATION_MESSAGE = 'Claude Code has switched from npm to native installer. Run `claude install` or see https://docs.anthropic.com/en/docs/claude-code/getting-started for more options.';
 export function useNpmDeprecationNotification() {
   useStartupNotification(_temp);
 }

src/hooks/useDiffInIDE.ts

@@ -60,7 +60,7 @@ export function useDiffInIDE({
 
   const sha = useMemo(() => randomUUID().slice(0, 6), [])
   const tabName = useMemo(
-    () => `✻ [OpenClaude] ${basename(filePath)} (${sha}) ⧉`,
+    () => `✻ [Claude Code] ${basename(filePath)} (${sha}) ⧉`,
     [filePath, sha],
   )
 

src/hooks/useVoice.ts

@@ -501,7 +501,7 @@ export function useVoice({
           } else if (!hadAudioSignal) {
             // Distinguish silent mic (capture issue) from speech not recognized.
             onErrorRef.current?.(
-              'No audio detected from microphone. Check that the correct input device is selected and that OpenClaude has microphone access.',
+              'No audio detected from microphone. Check that the correct input device is selected and that Claude Code has microphone access.',
             )
           } else {
             onErrorRef.current?.('No speech detected.')

src/main.tsx

@@ -785,7 +785,7 @@ export async function main() {
       // Headless (-p) mode is not supported with SSH in v1 — reject early
       // so the flag doesn't silently cause local execution.
       if (rest.includes('-p') || rest.includes('--print')) {
-        process.stderr.write('Error: headless (-p/--print) mode is not supported with openclaude ssh\n');
+        process.stderr.write('Error: headless (-p/--print) mode is not supported with claude ssh\n');
         gracefulShutdownSync(1);
         return;
       }
@@ -959,7 +959,7 @@ async function run(): Promise<CommanderCommand> {
     }
     profileCheckpoint('preAction_after_settings_sync');
   });
-  program.name('openclaude').description(`OpenClaude - starts an interactive session by default, use -p/--print for non-interactive output`).argument('[prompt]', 'Your prompt', String)
+  program.name('claude').description(`Claude Code - starts an interactive session by default, use -p/--print for non-interactive output`).argument('[prompt]', 'Your prompt', String)
   // Subcommands inherit helpOption via commander's copyInheritedSettings —
   // setting it once here covers mcp, plugin, auth, and all other subcommands.
   .helpOption('-h, --help', 'Display help for command').option('-d, --debug [filter]', 'Enable debug mode with optional category filtering (e.g., "api,hooks" or "!1p,!file")', (_value: string | true) => {
@@ -1013,7 +1013,7 @@ async function run(): Promise<CommanderCommand> {
     if (prompt === 'code') {
       logEvent('tengu_code_prompt_ignored', {});
       // biome-ignore lint/suspicious/noConsole:: intentional console output
-      console.warn(chalk.yellow('Tip: You can launch OpenClaude with just `openclaude`'));
+      console.warn(chalk.yellow('Tip: You can launch Claude Code with just `claude`'));
       prompt = undefined;
     }
 
@@ -3275,7 +3275,7 @@ async function run(): Promise<CommanderCommand> {
           }
           // The daemon needs a few seconds to spin up its worker and
           // establish a bridge session before discovery will find it.
-          return await exitWithMessage(root, `Assistant installed in ${installedDir}. The daemon is starting up — run \`openclaude assistant\` again in a few seconds to connect.`, {
+          return await exitWithMessage(root, `Assistant installed in ${installedDir}. The daemon is starting up — run \`claude assistant\` again in a few seconds to connect.`, {
             exitCode: 0,
             beforeExit: () => gracefulShutdown(0)
           });
@@ -3401,7 +3401,7 @@ async function run(): Promise<CommanderCommand> {
         // Check if TUI mode is enabled - description is only optional in TUI mode
         const isRemoteTuiEnabled = getFeatureValue_CACHED_MAY_BE_STALE('tengu_remote_backend', false);
         if (!isRemoteTuiEnabled && !hasInitialPrompt) {
-          return await exitWithError(root, 'Error: --remote requires a description.\nUsage: openclaude --remote "your task description"', () => gracefulShutdown(1));
+          return await exitWithError(root, 'Error: --remote requires a description.\nUsage: claude --remote "your task description"', () => gracefulShutdown(1));
         }
         logEvent('tengu_remote_create_session', {
           has_initial_prompt: String(hasInitialPrompt) as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS
@@ -3425,7 +3425,7 @@ async function run(): Promise<CommanderCommand> {
           // Original behavior: print session info and exit
           process.stdout.write(`Created remote session: ${createdSession.title}\n`);
           process.stdout.write(`View: ${getRemoteSessionUrl(createdSession.id)}?m=0\n`);
-          process.stdout.write(`Resume with: openclaude --teleport ${createdSession.id}\n`);
+          process.stdout.write(`Resume with: claude --teleport ${createdSession.id}\n`);
           await gracefulShutdown(0);
           process.exit(0);
         }
@@ -3537,7 +3537,7 @@ async function run(): Promise<CommanderCommand> {
                   }
                 } else {
                   // No known paths - show original error
-                  throw new TeleportOperationError(`You must run openclaude --teleport ${teleport} from a checkout of ${sessionRepo}.`, chalk.red(`You must run openclaude --teleport ${teleport} from a checkout of ${chalk.bold(sessionRepo)}.\n`));
+                  throw new TeleportOperationError(`You must run claude --teleport ${teleport} from a checkout of ${sessionRepo}.`, chalk.red(`You must run claude --teleport ${teleport} from a checkout of ${chalk.bold(sessionRepo)}.\n`));
                 }
               }
             } else if (repoValidation.status === 'error') {
@@ -3789,7 +3789,7 @@ async function run(): Promise<CommanderCommand> {
         pendingHookMessages
       }, renderAndRun);
     }
-  }).version(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (OpenClaude)`, '-v, --version', 'Output the version number');
+  }).version(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (Open Claude)`, '-v, --version', 'Output the version number');
 
   // Worktree flags
   program.option('-w, --worktree [name]', 'Create a new git worktree for this session (optionally specify a name)');
@@ -3876,7 +3876,7 @@ async function run(): Promise<CommanderCommand> {
   // claude mcp
 
   const mcp = program.command('mcp').description('Configure and manage MCP servers').configureHelp(createSortedHelpConfig()).enablePositionalOptions();
-  mcp.command('serve').description(`Start the OpenClaude MCP server`).option('-d, --debug', 'Enable debug mode', () => true).option('--verbose', 'Override verbose mode setting from config', () => true).action(async ({
+  mcp.command('serve').description(`Start the Claude Code MCP server`).option('-d, --debug', 'Enable debug mode', () => true).option('--verbose', 'Override verbose mode setting from config', () => true).action(async ({
     debug,
     verbose
   }: {
@@ -3944,7 +3944,7 @@ async function run(): Promise<CommanderCommand> {
 
   // claude server
   if (feature('DIRECT_CONNECT')) {
-    program.command('server').description('Start an OpenClaude session server').option('--port <number>', 'HTTP port', '0').option('--host <string>', 'Bind address', '0.0.0.0').option('--auth-token <token>', 'Bearer token for auth').option('--unix <path>', 'Listen on a unix domain socket').option('--workspace <dir>', 'Default working directory for sessions that do not specify cwd').option('--idle-timeout <ms>', 'Idle timeout for detached sessions in ms (0 = never expire)', '600000').option('--max-sessions <n>', 'Maximum concurrent sessions (0 = unlimited)', '32').action(async (opts: {
+    program.command('server').description('Start a Claude Code session server').option('--port <number>', 'HTTP port', '0').option('--host <string>', 'Bind address', '0.0.0.0').option('--auth-token <token>', 'Bearer token for auth').option('--unix <path>', 'Listen on a unix domain socket').option('--workspace <dir>', 'Default working directory for sessions that do not specify cwd').option('--idle-timeout <ms>', 'Idle timeout for detached sessions in ms (0 = never expire)', '600000').option('--max-sessions <n>', 'Maximum concurrent sessions (0 = unlimited)', '32').action(async (opts: {
       port: string;
       host: string;
       authToken?: string;
@@ -3978,7 +3978,7 @@ async function run(): Promise<CommanderCommand> {
       } = await import('./server/lockfile.js');
       const existing = await probeRunningServer();
       if (existing) {
-        process.stderr.write(`An OpenClaude server is already running (pid ${existing.pid}) at ${existing.httpUrl}\n`);
+        process.stderr.write(`A claude server is already running (pid ${existing.pid}) at ${existing.httpUrl}\n`);
         process.exit(1);
       }
       const authToken = opts.authToken ?? `sk-ant-cc-${randomBytes(16).toString('base64url')}`;
@@ -4028,11 +4028,11 @@ async function run(): Promise<CommanderCommand> {
   // this action it means the argv rewrite didn't fire (e.g. user ran
   // `claude ssh` with no host) — just print usage.
   if (feature('SSH_REMOTE')) {
-    program.command('ssh <host> [dir]').description('Run OpenClaude on a remote host over SSH. Deploys the binary and ' + 'tunnels API auth back through your local machine — no remote setup needed.').option('--permission-mode <mode>', 'Permission mode for the remote session').option('--dangerously-skip-permissions', 'Skip all permission prompts on the remote (dangerous)').option('--local', 'e2e test mode — spawn the child CLI locally (skip ssh/deploy). ' + 'Exercises the auth proxy and unix-socket plumbing without a remote host.').action(async () => {
+    program.command('ssh <host> [dir]').description('Run Claude Code on a remote host over SSH. Deploys the binary and ' + 'tunnels API auth back through your local machine — no remote setup needed.').option('--permission-mode <mode>', 'Permission mode for the remote session').option('--dangerously-skip-permissions', 'Skip all permission prompts on the remote (dangerous)').option('--local', 'e2e test mode — spawn the child CLI locally (skip ssh/deploy). ' + 'Exercises the auth proxy and unix-socket plumbing without a remote host.').action(async () => {
       // Argv rewriting in main() should have consumed `ssh <host>` before
       // commander runs. Reaching here means host was missing or the
       // rewrite predicate didn't match.
-      process.stderr.write('Usage: openclaude ssh <user@host | ssh-config-alias> [dir]\n\n' + "Runs OpenClaude on a remote Linux host. You don't need to install\n" + 'anything on the remote or run `openclaude auth login` there — the binary is\n' + 'deployed over SSH and API auth tunnels back through your local machine.\n');
+      process.stderr.write('Usage: claude ssh <user@host | ssh-config-alias> [dir]\n\n' + "Runs Claude Code on a remote Linux host. You don't need to install\n" + 'anything on the remote or run `claude auth login` there — the binary is\n' + 'deployed over SSH and API auth tunnels back through your local machine.\n');
       process.exit(1);
     });
   }
@@ -4041,7 +4041,7 @@ async function run(): Promise<CommanderCommand> {
   // Interactive mode (without -p) is handled by early argv rewriting in main()
   // which redirects to the main command with full TUI support.
   if (feature('DIRECT_CONNECT')) {
-    program.command('open <cc-url>').description('Connect to an OpenClaude server (internal — use cc:// URLs)').option('-p, --print [prompt]', 'Print mode (headless)').option('--output-format <format>', 'Output format: text, json, stream-json', 'text').action(async (ccUrl: string, opts: {
+    program.command('open <cc-url>').description('Connect to a Claude Code server (internal — use cc:// URLs)').option('-p, --print [prompt]', 'Print mode (headless)').option('--output-format <format>', 'Output format: text, json, stream-json', 'text').action(async (ccUrl: string, opts: {
       print?: string | boolean;
       outputFormat: string;
     }) => {
@@ -4130,7 +4130,7 @@ async function run(): Promise<CommanderCommand> {
   const coworkOption = () => new Option('--cowork', 'Use cowork_plugins directory').hideHelp();
 
   // Plugin validate command
-  const pluginCmd = program.command('plugin').alias('plugins').description('Manage OpenClaude plugins').configureHelp(createSortedHelpConfig());
+  const pluginCmd = program.command('plugin').alias('plugins').description('Manage Claude Code plugins').configureHelp(createSortedHelpConfig());
   pluginCmd.command('validate <path>').description('Validate a plugin or marketplace manifest').addOption(coworkOption()).action(async (manifestPath: string, options: {
     cowork?: boolean;
   }) => {
@@ -4153,7 +4153,7 @@ async function run(): Promise<CommanderCommand> {
   });
 
   // Marketplace subcommands
-  const marketplaceCmd = pluginCmd.command('marketplace').description('Manage OpenClaude marketplaces').configureHelp(createSortedHelpConfig());
+  const marketplaceCmd = pluginCmd.command('marketplace').description('Manage Claude Code marketplaces').configureHelp(createSortedHelpConfig());
   marketplaceCmd.command('add <source>').description('Add a marketplace from a URL, path, or GitHub repo').addOption(coworkOption()).option('--sparse <paths...>', 'Limit checkout to specific directories via git sparse-checkout (for monorepos). Example: --sparse .claude-plugin plugins').option('--scope <scope>', 'Where to declare the marketplace: user (default), project, or local').action(async (source: string, options: {
     cowork?: boolean;
     sparse?: string[];
@@ -4322,13 +4322,13 @@ async function run(): Promise<CommanderCommand> {
       // before commander runs. Reaching here means a root flag came first
       // (e.g. `--debug assistant`) and the position-0 predicate
       // didn't match. Print usage like the ssh stub does.
-      process.stderr.write('Usage: openclaude assistant [sessionId]\n\n' + 'Attach the REPL as a viewer client to a running bridge session.\n' + 'Omit sessionId to discover and pick from available sessions.\n');
+      process.stderr.write('Usage: claude assistant [sessionId]\n\n' + 'Attach the REPL as a viewer client to a running bridge session.\n' + 'Omit sessionId to discover and pick from available sessions.\n');
       process.exit(1);
     });
   }
 
   // Doctor command - check installation health
-  program.command('doctor').description('Check the health of your OpenClaude auto-updater. Note: The workspace trust dialog is skipped and stdio servers from .mcp.json are spawned for health checks. Only use this command in directories you trust.').action(async () => {
+  program.command('doctor').description('Check the health of your Claude Code auto-updater. Note: The workspace trust dialog is skipped and stdio servers from .mcp.json are spawned for health checks. Only use this command in directories you trust.').action(async () => {
     const [{
       doctorHandler
     }, {
@@ -4377,7 +4377,7 @@ async function run(): Promise<CommanderCommand> {
   }
 
   // claude install
-  program.command('install [target]').description('Install OpenClaude native build. Use [target] to specify version (stable, latest, or specific version)').option('--force', 'Force installation even if already installed').action(async (target: string | undefined, options: {
+  program.command('install [target]').description('Install Claude Code native build. Use [target] to specify version (stable, latest, or specific version)').option('--force', 'Force installation even if already installed').action(async (target: string | undefined, options: {
     force?: boolean;
   }) => {
     const {

src/query.ts

@@ -77,7 +77,6 @@ import {
 import { notifyCommandLifecycle } from './utils/commandLifecycle.js'
 import { headlessProfilerCheckpoint } from './utils/headlessProfiler.js'
 import {
-  getDefaultMainLoopModelSetting,
   getRuntimeMainLoopModel,
   renderModelName,
 } from './utils/model/model.js'
@@ -99,7 +98,6 @@ import { applyToolResultBudget } from './utils/toolResultStorage.js'
 import { recordContentReplacement } from './utils/sessionStorage.js'
 import { handleStopHooks } from './query/stopHooks.js'
 import { buildQueryConfig } from './query/config.js'
-import { getGlobalConfig } from './utils/config.js'
 import { productionDeps, type QueryDeps } from './query/deps.js'
 import type { Terminal, Continue } from './query/transitions.js'
 import { feature } from 'bun:bundle'
@@ -110,6 +108,7 @@ import {
 } from './bootstrap/state.js'
 import { createBudgetTracker, checkTokenBudget } from './query/tokenBudget.js'
 import { count } from './utils/array.js'
+
 /* eslint-disable @typescript-eslint/no-require-imports */
 const snipModule = feature('HISTORY_SNIP')
   ? (require('./services/compact/snipCompact.js') as typeof import('./services/compact/snipCompact.js'))
@@ -253,15 +252,6 @@ async function* queryLoop(
   | ToolUseSummaryMessage,
   Terminal
 > {
-  // Start a new turn for multi-turn context tracking
-  if (
-    feature('MULTI_TURN_CONTEXT') &&
-    getGlobalConfig().knowledgeGraphEnabled
-  ) {
-    const { startNewTurn } = await import('./utils/multiTurnContext.js')
-    startNewTurn()
-  }
-
   // Immutable params — never reassigned during the query loop.
   const {
     systemPrompt,
@@ -378,16 +368,6 @@ async function* queryLoop(
 
     let messagesForQuery = [...getMessagesAfterCompactBoundary(messages)]
 
-    // Extract facts and update phase from the latest message (user input or tool result)
-    if (
-      feature('CONVERSATION_ARC') &&
-      getGlobalConfig().knowledgeGraphEnabled &&
-      messagesForQuery.length > 0
-    ) {
-      const { updateArcPhase } = await import('./utils/conversationArc.js')
-      updateArcPhase([messagesForQuery[messagesForQuery.length - 1]])
-    }
-
     let tracking = autoCompactTracking
 
     // Enforce per-message budget on aggregate tool result size. Runs BEFORE
@@ -476,27 +456,8 @@ async function* queryLoop(
       messagesForQuery = collapseResult.messages
     }
 
-    // arcSummary must be a separate array element; concatenating it into a
-    // template string makes [...systemPrompt] spread chars, shredding the prompt.
-    let promptWithArc: readonly string[] = systemPrompt
-    if (feature('CONVERSATION_ARC')) {
-      if (getGlobalConfig().knowledgeGraphEnabled) {
-        const lastMessage = messagesForQuery[messagesForQuery.length - 1]
-        const userQueryText =
-          lastMessage?.type === 'user' &&
-          typeof lastMessage.message.content === 'string'
-            ? lastMessage.message.content
-            : ''
-        const { getArcSummary } = await import('./utils/conversationArc.js')
-        const arcSummary = getArcSummary(userQueryText)
-        if (arcSummary) {
-          promptWithArc = [...systemPrompt, arcSummary]
-        }
-      }
-    }
-
     const fullSystemPrompt = asSystemPrompt(
-      appendSystemContext(asSystemPrompt(promptWithArc), systemContext),
+      appendSystemContext(systemPrompt, systemContext),
     )
 
     queryCheckpoint('query_autocompact_start')
@@ -618,13 +579,9 @@ async function* queryLoop(
 
     const appState = toolUseContext.getAppState()
     const permissionMode = appState.toolPermissionContext.mode
-    const appStateMainLoopModel =
-      appState.mainLoopModelForSession ??
-      appState.mainLoopModel ??
-      getDefaultMainLoopModelSetting()
     let currentModel = getRuntimeMainLoopModel({
       permissionMode,
-      mainLoopModel: appStateMainLoopModel,
+      mainLoopModel: toolUseContext.options.mainLoopModel,
       exceeds200kTokens:
         permissionMode === 'plan' &&
         doesMostRecentAssistantMessageExceed200k(messagesForQuery),
@@ -1559,34 +1516,6 @@ async function* queryLoop(
     }
     queryCheckpoint('query_tool_execution_end')
 
-    // Track multi-turn context after tool execution
-    if (
-      feature('MULTI_TURN_CONTEXT') &&
-      getGlobalConfig().knowledgeGraphEnabled
-    ) {
-      const { addMessageToTurn, addToolCallToTurn } = await import(
-        './utils/multiTurnContext.js'
-      )
-      addMessageToTurn(assistantMessage)
-      for (const toolUse of toolUseBlocks) {
-        addToolCallToTurn({
-          id: toolUse.id,
-          name: toolUse.name,
-          input: toolUse.input as Record<string, unknown>,
-          timestamp: Date.now(),
-        })
-      }
-    }
-
-    // Update conversation arc phase
-    if (
-      feature('CONVERSATION_ARC') &&
-      getGlobalConfig().knowledgeGraphEnabled
-    ) {
-      const { updateArcPhase } = await import('./utils/conversationArc.js')
-      updateArcPhase([assistantMessage])
-    }
-
     // Generate tool use summary after tool batch completes — passed to next recursive call
     let nextPendingToolUseSummary:
       | Promise<ToolUseSummaryMessage | null>
@@ -1891,15 +1820,6 @@ async function* queryLoop(
     }
 
     queryCheckpoint('query_recursive_call')
-
-    if (
-      feature('CONVERSATION_ARC') &&
-      getGlobalConfig().knowledgeGraphEnabled
-    ) {
-      const { finalizeArcTurn } = await import('./utils/conversationArc.js')
-      finalizeArcTurn()
-    }
-
     const next: State = {
       messages: [...messagesForQuery, ...assistantMessages, ...toolResults],
       toolUseContext: toolUseContextWithQueryTracking,

src/screens/Doctor.tsx

@@ -222,7 +222,7 @@ export function Doctor(t0) {
   let t7;
   if ($[11] !== onDone) {
     t7 = () => {
-      onDone("OpenClaude diagnostics dismissed", {
+      onDone("Claude Code diagnostics dismissed", {
         display: "system"
       });
     };

src/screens/REPL.tsx

@@ -133,8 +133,6 @@ import { hasConsoleBillingAccess } from '../utils/billing.js';
 import { logEvent, type AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS } from 'src/services/analytics/index.js';
 import { getFeatureValue_CACHED_MAY_BE_STALE } from 'src/services/analytics/growthbook.js';
 import { textForResubmit, handleMessageFromStream, type StreamingToolUse, type StreamingThinking, isCompactBoundaryMessage, getMessagesAfterCompactBoundary, getContentText, createUserMessage, createAssistantMessage, createTurnDurationMessage, createAgentsKilledMessage, createApiMetricsMessage, createSystemMessage, createCommandInputMessage, formatCommandInputTags } from '../utils/messages.js';
-import { getCurrentTurnCacheMetrics, resetCurrentTurn } from '../services/api/cacheStatsTracker.js';
-import { formatCacheMetricsCompact, formatCacheMetricsFull } from '../services/api/cacheMetrics.js';
 import { generateSessionTitle } from '../utils/sessionTitle.js';
 import { BASH_INPUT_TAG, COMMAND_MESSAGE_TAG, COMMAND_NAME_TAG, LOCAL_COMMAND_STDOUT_TAG } from '../constants/xml.js';
 import { escapeXml } from '../utils/xml.js';
@@ -1136,7 +1134,7 @@ export function REPL({
   // session from mid-conversation context.
   const haikuTitleAttemptedRef = useRef((initialMessages?.length ?? 0) > 0);
   const agentTitle = mainThreadAgentDefinition?.agentType;
-  const terminalTitle = sessionTitle ?? agentTitle ?? haikuTitle ?? 'OpenClaude';
+  const terminalTitle = sessionTitle ?? agentTitle ?? haikuTitle ?? 'Open Claude';
   const isWaitingForApproval = toolUseConfirmQueue.length > 0 || promptQueue.length > 0 || pendingWorkerRequest || pendingSandboxRequest;
   // Local-jsx commands (like /plugin, /config) show user-facing dialogs that
   // wait for input. Require jsx != null — if the flag is stuck true but jsx
@@ -2923,13 +2921,6 @@ export function REPL({
       // isLoading is derived from queryGuard — tryStart() above already
       // transitioned dispatching→running, so no setter call needed here.
       resetTimingRefs();
-      // Start-of-turn cache tracker reset. The end-of-turn path at the
-      // bottom of this function already resets, but mirror the call here
-      // so a turn that never reaches end-of-turn (crash, unhandled
-      // rejection, process exit) still starts clean on the next one.
-      // Idempotent with respect to the end-of-turn reset — double-reset
-      // is a no-op.
-      resetCurrentTurn();
       setMessages(oldMessages => [...oldMessages, ...newMessages]);
       responseLengthRef.current = 0;
       if (feature('TOKEN_BUDGET')) {
@@ -3028,38 +3019,6 @@ export function REPL({
             setMessages(prev => [...prev, createTurnDurationMessage(turnDurationMs, budgetInfo, count(prev, isLoggableMessage))]);
           }
         }
-        // Cache stats line — controlled by `/config showCacheStats`. Shows
-        // per-query read/hit stats using the provider-normalized metrics
-        // from cacheStatsTracker. 'off' skips, 'compact' gives a one-liner,
-        // 'full' gives a breakdown. Display is skipped when the user
-        // aborted or proactive mode is active — but the counter reset
-        // below still runs in those cases.
-        if (!abortController.signal.aborted && !proactiveActive) {
-          // Defensive default: config layer already merges 'compact' from
-          // DEFAULT_GLOBAL_CONFIG (see config.ts:1494) for configs that
-          // predate this feature, so `mode` should always be defined.
-          // The `?? 'compact'` fallback covers pathological cases — a
-          // corrupt config read that returned an empty object, or a
-          // race between writer and reader — where the merge didn't
-          // land. Rendering the line is the safer failure mode than
-          // silently hiding it.
-          const mode = getGlobalConfig().showCacheStats ?? 'compact';
-          if (mode !== 'off') {
-            const turnMetrics = getCurrentTurnCacheMetrics();
-            // Skip rendering if the turn recorded no API activity at all —
-            // avoids a spurious "[Cache: cold]" on local-only commands.
-            if (turnMetrics.supported || turnMetrics.read > 0 || turnMetrics.total > 0) {
-              const line = mode === 'full' ? formatCacheMetricsFull(turnMetrics) : formatCacheMetricsCompact(turnMetrics);
-              setMessages(prev => [...prev, createSystemMessage(line, 'info')]);
-            }
-          }
-        }
-        // Reset turn counters UNCONDITIONALLY — users routinely interrupt
-        // (Ctrl+C) mid-turn, and if we kept the reset gated on
-        // !aborted, the in-flight turn's metrics would leak into the
-        // next turn's aggregate. Proactive turns also need the reset so
-        // their metrics don't pile onto the following user turn.
-        resetCurrentTurn();
         // Clear the controller so CancelRequestHandler's canCancelRunningTask
         // reads false at the idle prompt. Without this, the stale non-aborted
         // controller makes ctrl+c fire onCancel() (aborting nothing) instead of
@@ -4189,7 +4148,7 @@ export function REPL({
   useEffect(() => {
     const handleSuspend = () => {
       // Print suspension instructions
-      process.stdout.write(`\nOpenClaude has been suspended. Run \`fg\` to bring OpenClaude back.\nNote: ctrl + z now suspends OpenClaude, ctrl + _ undoes input.\n`);
+      process.stdout.write(`\nClaude Code has been suspended. Run \`fg\` to bring Claude Code back.\nNote: ctrl + z now suspends Claude Code, ctrl + _ undoes input.\n`);
     };
     const handleResume = () => {
       // Force complete component tree replacement instead of terminal clear

src/services/api/__fixtures__/minimax-model-remains.json

@@ -1,21 +0,0 @@
-{
-  "plan_type": "plus_highspeed",
-  "model_remains": [
-    {
-      "model_name": "MiniMax-M2.7",
-      "remains_time": 3600,
-      "current_interval_total_count": 1500,
-      "current_interval_usage_count": 1437
-    },
-    {
-      "model_name": "MiniMax-M2.7-highspeed",
-      "end_time": 1771603200000,
-      "current_interval_total_count": 2000,
-      "current_interval_usage_count": 1000
-    }
-  ],
-  "base_resp": {
-    "status_code": 0,
-    "status_msg": "success"
-  }
-}

src/services/api/cacheMetrics.test.ts

@@ -1,782 +0,0 @@
-import { expect, test, describe } from 'bun:test'
-import {
-  extractCacheMetrics,
-  extractCacheReadFromRawUsage,
-  resolveCacheProvider,
-  formatCacheMetricsCompact,
-  formatCacheMetricsFull,
-  addCacheMetrics,
-} from './cacheMetrics.js'
-
-describe('extractCacheMetrics — Anthropic (firstParty/bedrock/vertex/foundry)', () => {
-  test('reports read/created separately and computes hit rate over total input', () => {
-    const usage = {
-      input_tokens: 300,
-      output_tokens: 100,
-      cache_read_input_tokens: 800,
-      cache_creation_input_tokens: 200,
-    }
-    const m = extractCacheMetrics(usage, 'anthropic')
-    expect(m.supported).toBe(true)
-    expect(m.read).toBe(800)
-    expect(m.created).toBe(200)
-    // total = fresh(300) + created(200) + read(800) = 1300
-    expect(m.total).toBe(1300)
-    expect(m.hitRate).toBeCloseTo(800 / 1300, 4)
-  })
-
-  test('returns cold metrics when no cache activity yet', () => {
-    const m = extractCacheMetrics({ input_tokens: 500 }, 'anthropic')
-    expect(m.supported).toBe(true)
-    expect(m.read).toBe(0)
-    expect(m.created).toBe(0)
-    expect(m.hitRate).toBe(0)
-  })
-
-  test('null hit rate when usage has no input at all', () => {
-    const m = extractCacheMetrics({}, 'anthropic')
-    expect(m.supported).toBe(true)
-    expect(m.hitRate).toBeNull()
-  })
-})
-
-// NOTE: OpenAI/Codex/Kimi/DeepSeek/Gemini raw shapes are now tested through
-// extractCacheReadFromRawUsage (below). extractCacheMetrics sees the
-// post-shim Anthropic shape for every provider, so the tests here verify
-// that the shape lookup works uniformly against the shimmed fields.
-
-describe('extractCacheMetrics — post-shim Anthropic shape (applies to all providers)', () => {
-  test('OpenAI post-shim (openai bucket) — reads Anthropic fields injected by convertChunkUsage', () => {
-    // This is what cost-tracker actually sees for OpenAI upstreams: the
-    // shim has already subtracted cached from prompt_tokens and moved it
-    // to cache_read_input_tokens.
-    const shimmed = {
-      input_tokens: 800, // fresh = 2000 - 1200
-      output_tokens: 300,
-      cache_creation_input_tokens: 0,
-      cache_read_input_tokens: 1_200,
-    }
-    const m = extractCacheMetrics(shimmed, 'openai')
-    expect(m.supported).toBe(true)
-    expect(m.read).toBe(1_200)
-    expect(m.created).toBe(0)
-    expect(m.total).toBe(2_000) // 800 fresh + 1200 read
-    expect(m.hitRate).toBe(0.6)
-  })
-
-  test('Codex post-shim — same Anthropic shape as OpenAI', () => {
-    const shimmed = {
-      input_tokens: 900, // 1500 - 600
-      cache_creation_input_tokens: 0,
-      cache_read_input_tokens: 600,
-    }
-    const m = extractCacheMetrics(shimmed, 'codex')
-    expect(m.read).toBe(600)
-    expect(m.total).toBe(1_500)
-    expect(m.hitRate).toBe(0.4)
-  })
-
-  test('Kimi post-shim — shim moved top-level cached_tokens into Anthropic field', () => {
-    const shimmed = {
-      input_tokens: 600, // 1000 - 400
-      cache_creation_input_tokens: 0,
-      cache_read_input_tokens: 400,
-    }
-    const m = extractCacheMetrics(shimmed, 'kimi')
-    expect(m.read).toBe(400)
-    expect(m.total).toBe(1_000)
-    expect(m.hitRate).toBe(0.4)
-  })
-
-  test('DeepSeek post-shim — hit moved to cache_read_input_tokens, miss to input_tokens', () => {
-    const shimmed = {
-      input_tokens: 300, // miss
-      cache_creation_input_tokens: 0,
-      cache_read_input_tokens: 700, // hit
-    }
-    const m = extractCacheMetrics(shimmed, 'deepseek')
-    expect(m.read).toBe(700)
-    expect(m.total).toBe(1_000)
-    expect(m.hitRate).toBe(0.7)
-  })
-
-  test('Gemini post-shim — cached_content_token_count moved to cache_read_input_tokens', () => {
-    const shimmed = {
-      input_tokens: 800, // 4000 - 3200
-      cache_creation_input_tokens: 0,
-      cache_read_input_tokens: 3_200,
-    }
-    const m = extractCacheMetrics(shimmed, 'gemini')
-    expect(m.read).toBe(3_200)
-    expect(m.total).toBe(4_000)
-    expect(m.hitRate).toBe(0.8)
-  })
-})
-
-describe('extractCacheReadFromRawUsage — single source of truth for shim layer', () => {
-  test('Anthropic-native passthrough: cache_read_input_tokens', () => {
-    expect(
-      extractCacheReadFromRawUsage({ cache_read_input_tokens: 1_500 }),
-    ).toBe(1_500)
-  })
-
-  test('OpenAI Chat Completions: prompt_tokens_details.cached_tokens', () => {
-    expect(
-      extractCacheReadFromRawUsage({
-        prompt_tokens: 2_000,
-        prompt_tokens_details: { cached_tokens: 1_200 },
-      }),
-    ).toBe(1_200)
-  })
-
-  test('Codex Responses API: input_tokens_details.cached_tokens', () => {
-    expect(
-      extractCacheReadFromRawUsage({
-        input_tokens: 1_500,
-        input_tokens_details: { cached_tokens: 600 },
-      }),
-    ).toBe(600)
-  })
-
-  test('Kimi / Moonshot: top-level cached_tokens', () => {
-    expect(
-      extractCacheReadFromRawUsage({ prompt_tokens: 1_000, cached_tokens: 400 }),
-    ).toBe(400)
-  })
-
-  test('DeepSeek: prompt_cache_hit_tokens', () => {
-    expect(
-      extractCacheReadFromRawUsage({
-        prompt_cache_hit_tokens: 700,
-        prompt_cache_miss_tokens: 300,
-      }),
-    ).toBe(700)
-  })
-
-  test('Gemini: cached_content_token_count', () => {
-    expect(
-      extractCacheReadFromRawUsage({
-        prompt_token_count: 4_000,
-        cached_content_token_count: 3_200,
-      }),
-    ).toBe(3_200)
-  })
-
-  test('no cache fields at all → 0 (Copilot/Ollama/unknown shape)', () => {
-    expect(extractCacheReadFromRawUsage({ prompt_tokens: 500 })).toBe(0)
-  })
-
-  test('Anthropic field wins over OpenAI field when both present', () => {
-    // Shouldn't happen in practice, but if usage was double-annotated we
-    // trust the Anthropic-native number (it's the more authoritative one).
-    expect(
-      extractCacheReadFromRawUsage({
-        cache_read_input_tokens: 999,
-        prompt_tokens_details: { cached_tokens: 111 },
-      }),
-    ).toBe(999)
-  })
-
-  test('null/undefined/non-object → 0', () => {
-    expect(extractCacheReadFromRawUsage(null)).toBe(0)
-    expect(extractCacheReadFromRawUsage(undefined)).toBe(0)
-    expect(extractCacheReadFromRawUsage('nope' as unknown as never)).toBe(0)
-  })
-})
-
-describe('extractCacheMetrics — Copilot / Ollama (unsupported)', () => {
-  test('returns supported:false with all zeros and null hitRate for Copilot', () => {
-    const m = extractCacheMetrics({ prompt_tokens: 1000 }, 'copilot')
-    expect(m.supported).toBe(false)
-    expect(m.read).toBe(0)
-    expect(m.created).toBe(0)
-    expect(m.hitRate).toBeNull()
-  })
-
-  test('returns supported:false for Ollama', () => {
-    const m = extractCacheMetrics({ prompt_tokens: 42 }, 'ollama')
-    expect(m.supported).toBe(false)
-    expect(m.hitRate).toBeNull()
-  })
-
-  test('Copilot serving Claude (copilot-claude) is supported and uses Anthropic fields', () => {
-    const usage = {
-      input_tokens: 200,
-      cache_read_input_tokens: 800,
-      cache_creation_input_tokens: 100,
-    }
-    const m = extractCacheMetrics(usage, 'copilot-claude')
-    expect(m.supported).toBe(true)
-    expect(m.read).toBe(800)
-    expect(m.created).toBe(100)
-    expect(m.total).toBe(1_100)
-  })
-})
-
-describe('extractCacheMetrics — bad/empty input', () => {
-  test('null usage returns unsupported', () => {
-    expect(extractCacheMetrics(null, 'anthropic').supported).toBe(false)
-  })
-
-  test('non-object usage returns unsupported', () => {
-    expect(extractCacheMetrics('oops' as unknown as never, 'openai').supported).toBe(
-      false,
-    )
-  })
-})
-
-describe('resolveCacheProvider', () => {
-  test('firstParty → anthropic', () => {
-    expect(resolveCacheProvider('firstParty')).toBe('anthropic')
-  })
-  test('bedrock/vertex/foundry → anthropic', () => {
-    expect(resolveCacheProvider('bedrock')).toBe('anthropic')
-    expect(resolveCacheProvider('vertex')).toBe('anthropic')
-    expect(resolveCacheProvider('foundry')).toBe('anthropic')
-  })
-  test('github without claude hint → copilot (unsupported)', () => {
-    expect(resolveCacheProvider('github')).toBe('copilot')
-  })
-  test('github with claude hint → copilot-claude', () => {
-    expect(
-      resolveCacheProvider('github', { githubNativeAnthropic: true }),
-    ).toBe('copilot-claude')
-  })
-  test('openai with localhost / loopback → self-hosted', () => {
-    // These used to return 'ollama'; the bucket is now 'self-hosted'
-    // because not every local OpenAI-compatible server is Ollama
-    // (could be vLLM, LM Studio, LocalAI, text-generation-webui).
-    // Both buckets collapse to supported=false downstream.
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://localhost:8080/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://127.0.0.1:1234/v1' }),
-    ).toBe('self-hosted')
-    // Localhost:11434 hits the self-hosted branch first — 'ollama' only
-    // kicks in when the :11434 port appears on a public-looking URL
-    // (which would be unusual but still deserves honest classification).
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://localhost:11434/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://[::1]:5000/v1' }),
-    ).toBe('self-hosted')
-  })
-
-  test('openai on RFC1918 private IP → self-hosted (pre-fix: misclassified as openai)', () => {
-    // These are the exact cases the reviewer flagged. Before this fix,
-    // a vLLM / LocalAI server on a LAN address fell through to the
-    // 'openai' branch and /cache-stats showed '[Cache: cold]' — which
-    // users read as "my cache is broken" when the provider simply
-    // didn't report cache fields. Now they land in 'self-hosted' and
-    // /cache-stats shows '[Cache: N/A]'.
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://192.168.1.50:8000/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://10.0.0.7:8080/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://172.20.0.3:5000/v1' }),
-    ).toBe('self-hosted')
-  })
-
-  test('openai on link-local / CGNAT → self-hosted', () => {
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://169.254.169.254/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://100.64.1.5:8000/v1' }),
-    ).toBe('self-hosted')
-  })
-
-  test('openai on reserved TLD (.local / .internal / .lan / .home.arpa) → self-hosted', () => {
-    // Per RFC 6761 (.local/mDNS), RFC 8375 (.home.arpa), and widely
-    // used .internal / .lan conventions. These never resolve publicly.
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://llm.internal:5000/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://llm.local:8080/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://vllm.home.arpa/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://box.lan:1234/v1' }),
-    ).toBe('self-hosted')
-  })
-
-  test('openai on IPv6 local / link-local → self-hosted', () => {
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://[fe80::1]:8000/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://[fd12:3456::7]:8080/v1' }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'http://[fc00::1]:8080/v1' }),
-    ).toBe('self-hosted')
-  })
-
-  test('IPv6 ULA prefix (fc/fd) does NOT over-match public hostnames', () => {
-    // Regression guard: an early version of isLocalOrPrivateUrl checked
-    // `h.startsWith('fc')` / `startsWith('fd')` without a colon guard,
-    // which misclassified legitimate public hosts whose names happen to
-    // begin with those letters. The fix requires a colon in the match
-    // so only real IPv6 literals hit the branch.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://fc-api.example.com/v1',
-      }),
-    ).toBe('openai')
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://fd-hosted.example.com/v1',
-      }),
-    ).toBe('openai')
-    // Same goes for names that look like hex prefixes but aren't IPv6.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://fcbench.net/v1',
-      }),
-    ).toBe('openai')
-  })
-
-  test('openai with :11434 on a public host → ollama (default-port heuristic)', () => {
-    // Contrived but the heuristic should still fire — someone running
-    // Ollama behind a reverse proxy with port preserved.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://ollama.example.com:11434/v1',
-      }),
-    ).toBe('ollama')
-  })
-
-  test('openai with moonshot URL → kimi', () => {
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'https://api.moonshot.ai/v1' }),
-    ).toBe('kimi')
-  })
-  test('openai with deepseek URL → deepseek', () => {
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'https://api.deepseek.com/v1' }),
-    ).toBe('deepseek')
-  })
-  test('private IP beats hosted-keyword matching (self-hosted takes priority)', () => {
-    // A pathological URL: a private-IP host whose path string contains
-    // "deepseek". Self-hosted detection must run FIRST so the URL
-    // classifies honestly — the path alone doesn't prove the upstream
-    // is the real DeepSeek API.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'http://10.0.0.5:8000/v1/deepseek-proxy',
-      }),
-    ).toBe('self-hosted')
-  })
-  test('plain openai remains openai', () => {
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'https://api.openai.com/v1' }),
-    ).toBe('openai')
-  })
-  test('unparseable base URL falls back to substring heuristic', () => {
-    // Bare host:port without a scheme is common in misconfigured env.
-    // We can't URL-parse it, but we still honor the "localhost" hint so
-    // a broken config doesn't silently masquerade as cache-capable.
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: 'localhost:8000' }),
-    ).toBe('self-hosted')
-    // An unparseable and opaque string falls through to plain 'openai'
-    // (best-effort — nothing we can infer from "foo-bar-baz").
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: '???' }),
-    ).toBe('openai')
-  })
-  test('empty base URL → plain openai', () => {
-    // No hint at all: assume the canonical api.openai.com.
-    expect(resolveCacheProvider('openai')).toBe('openai')
-    expect(
-      resolveCacheProvider('openai', { openAiBaseUrl: '' }),
-    ).toBe('openai')
-  })
-  test('codex → codex', () => {
-    expect(resolveCacheProvider('codex')).toBe('codex')
-  })
-  test('gemini → gemini', () => {
-    expect(resolveCacheProvider('gemini')).toBe('gemini')
-  })
-})
-
-describe('resolveCacheProvider — .localhost TLD (RFC 6761)', () => {
-  test('subdomains of .localhost classify as self-hosted', () => {
-    // Chrome, Firefox, and systemd-resolved all natively resolve
-    // *.localhost to 127.0.0.1. Kubernetes Ingress and docker-compose
-    // setups commonly use app.localhost, api.localhost, etc.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'http://app.localhost:3000/v1',
-      }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'http://api.localhost/v1',
-      }),
-    ).toBe('self-hosted')
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'http://llm.dev.localhost:8080/v1',
-      }),
-    ).toBe('self-hosted')
-  })
-
-  test('.localhost TLD does NOT match substring collisions', () => {
-    // Guard against regressions where `localhost` would match via
-    // substring rather than TLD semantics. `localhostify.com` and
-    // `mylocalhost.net` must stay on the public `openai` path.
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://localhostify.com/v1',
-      }),
-    ).toBe('openai')
-    expect(
-      resolveCacheProvider('openai', {
-        openAiBaseUrl: 'https://mylocalhost.net/v1',
-      }),
-    ).toBe('openai')
-  })
-})
-
-describe('extractCacheMetrics — hit rate clamp', () => {
-  test('hitRate is clamped to 1.0 on pathological input (read > total)', () => {
-    // Defensive guard: with valid non-negative inputs the math enforces
-    // read <= total, so hitRate cannot exceed 1. But an upstream shim
-    // bug (e.g. reading a negative `fresh` from a future provider) could
-    // break the invariant. `Math.min(1, read/total)` caps the display at
-    // 100% rather than letting a `read=800 total=500` case render as
-    // "hit 160%" or (worse) null, which would hide the anomaly.
-    const metrics = extractCacheMetrics(
-      {
-        cache_read_input_tokens: 800,
-        cache_creation_input_tokens: 0,
-        // asNumber keeps finite negatives, so fresh = -500 → total =
-        // 800 + 0 + (-500) = 300, read=800 → raw ratio 2.67, clamp to 1.
-        input_tokens: -500,
-      } as unknown as Record<string, unknown>,
-      'anthropic',
-    )
-    expect(metrics.supported).toBe(true)
-    expect(metrics.hitRate).toBe(1)
-  })
-
-  test('normal inputs still yield accurate fractional hit rates', () => {
-    // Regression: clamp must not perturb the happy path.
-    const metrics = extractCacheMetrics(
-      {
-        cache_read_input_tokens: 300,
-        cache_creation_input_tokens: 0,
-        input_tokens: 700,
-      },
-      'anthropic',
-    )
-    expect(metrics.hitRate).toBeCloseTo(0.3, 5)
-  })
-})
-
-describe('extractCacheMetrics — self-hosted bucket (data-driven)', () => {
-  test('vanilla self-hosted endpoint without cache fields → unsupported / N/A', () => {
-    // vLLM, LocalAI, text-generation-webui, etc. emit no cache fields
-    // at all. With read=created=0 we mark unsupported so the REPL shows
-    // honest '[Cache: N/A]' instead of a fabricated 0%.
-    const metrics = extractCacheMetrics(
-      { input_tokens: 1_000, output_tokens: 200 },
-      'self-hosted',
-    )
-    expect(metrics.supported).toBe(false)
-    expect(metrics.hitRate).toBeNull()
-    expect(metrics.read).toBe(0)
-    expect(metrics.created).toBe(0)
-  })
-
-  test('internal reverse proxy forwarding real cache data → supported', () => {
-    // Review-blocker regression guard: an enterprise setup with an
-    // internal proxy on a private URL (e.g. `http://llm.internal:5000/v1`)
-    // forwarding to OpenAI / Kimi / DeepSeek / Gemini WILL deliver real
-    // cache fields via the shim. Pre-fix we would discard them because
-    // the URL heuristic classified the endpoint as 'self-hosted'. Now
-    // the data itself decides: any non-zero cache activity flows through
-    // the same normalization as an OpenAI bucket.
-    const shimmed = {
-      input_tokens: 800, // fresh (post-shim, cached already subtracted)
-      cache_read_input_tokens: 1_200, // shim extracted from upstream
-      cache_creation_input_tokens: 0,
-    }
-    const metrics = extractCacheMetrics(shimmed, 'self-hosted')
-    expect(metrics.supported).toBe(true)
-    expect(metrics.read).toBe(1_200)
-    expect(metrics.total).toBe(2_000)
-    expect(metrics.hitRate).toBe(0.6)
-  })
-
-  test('proxy with cache_creation but zero cache_read → still supported', () => {
-    // Mirror of the above for the first-call / cold-cache scenario:
-    // Anthropic-compatible upstreams emit creation tokens on the first
-    // request that primes the cache. Self-hosted proxy must preserve
-    // that signal, not swallow it because read is still 0.
-    const shimmed = {
-      input_tokens: 500,
-      cache_read_input_tokens: 0,
-      cache_creation_input_tokens: 800,
-    }
-    const metrics = extractCacheMetrics(shimmed, 'self-hosted')
-    expect(metrics.supported).toBe(true)
-    expect(metrics.created).toBe(800)
-    expect(metrics.read).toBe(0)
-  })
-})
-
-describe('formatCacheMetrics — defensive null/undefined guards', () => {
-  test('formatCacheMetricsCompact returns N/A for undefined input', () => {
-    // Signature says `CacheMetrics` but runtime bug on a failed API
-    // response could leave the caller with nothing. The formatter
-    // should degrade gracefully rather than throw on `.supported`.
-    expect(formatCacheMetricsCompact(undefined)).toBe('[Cache: N/A]')
-    expect(formatCacheMetricsCompact(null as unknown as undefined)).toBe(
-      '[Cache: N/A]',
-    )
-  })
-
-  test('formatCacheMetricsFull returns N/A for undefined input', () => {
-    expect(formatCacheMetricsFull(undefined)).toBe('[Cache: N/A]')
-    expect(formatCacheMetricsFull(null as unknown as undefined)).toBe(
-      '[Cache: N/A]',
-    )
-  })
-})
-
-describe('formatCacheMetricsCompact — self-hosted display paths', () => {
-  test('vanilla self-hosted (no cache data) renders as N/A', () => {
-    const metrics = extractCacheMetrics(
-      { input_tokens: 500 },
-      'self-hosted',
-    )
-    expect(formatCacheMetricsCompact(metrics)).toBe('[Cache: N/A]')
-    expect(formatCacheMetricsFull(metrics)).toBe('[Cache: N/A]')
-  })
-
-  test('self-hosted proxy with forwarded cache data renders real metrics', () => {
-    // Full display-path regression guard for the review-blocker fix:
-    // the user must see the real hit rate that the upstream emitted,
-    // not a silent N/A because the URL looked private.
-    const metrics = extractCacheMetrics(
-      {
-        input_tokens: 800,
-        cache_read_input_tokens: 1_200,
-        cache_creation_input_tokens: 0,
-      },
-      'self-hosted',
-    )
-    expect(formatCacheMetricsCompact(metrics)).toBe('[Cache: 1.2k read • hit 60%]')
-    expect(formatCacheMetricsFull(metrics)).toBe(
-      '[Cache: read=1.2k created=0 hit=60%]',
-    )
-  })
-})
-
-describe('formatCacheMetricsCompact — snapshot-stable output', () => {
-  test('supported with reads shows "k" abbreviation and hit rate', () => {
-    const out = formatCacheMetricsCompact({
-      read: 1_234,
-      created: 0,
-      total: 10_000,
-      hitRate: 0.1234,
-      supported: true,
-    })
-    expect(out).toBe('[Cache: 1.2k read • hit 12%]')
-  })
-
-  test('supported with no cache activity renders "cold"', () => {
-    const out = formatCacheMetricsCompact({
-      read: 0,
-      created: 0,
-      total: 500,
-      hitRate: 0,
-      supported: true,
-    })
-    expect(out).toBe('[Cache: cold]')
-  })
-
-  test('unsupported renders "N/A"', () => {
-    const out = formatCacheMetricsCompact({
-      read: 0,
-      created: 0,
-      total: 0,
-      hitRate: null,
-      supported: false,
-    })
-    expect(out).toBe('[Cache: N/A]')
-  })
-
-  test('small numbers render without abbreviation', () => {
-    const out = formatCacheMetricsCompact({
-      read: 42,
-      created: 0,
-      total: 100,
-      hitRate: 0.42,
-      supported: true,
-    })
-    expect(out).toBe('[Cache: 42 read • hit 42%]')
-  })
-})
-
-describe('formatCacheMetricsFull — snapshot-stable output', () => {
-  test('supported shows all fields', () => {
-    const out = formatCacheMetricsFull({
-      read: 1_234,
-      created: 250,
-      total: 10_000,
-      hitRate: 0.1234,
-      supported: true,
-    })
-    expect(out).toBe('[Cache: read=1.2k created=250 hit=12%]')
-  })
-
-  test('null hit rate renders n/a', () => {
-    const out = formatCacheMetricsFull({
-      read: 0,
-      created: 0,
-      total: 0,
-      hitRate: null,
-      supported: true,
-    })
-    expect(out).toBe('[Cache: read=0 created=0 hit=n/a]')
-  })
-
-  test('unsupported renders "N/A"', () => {
-    const out = formatCacheMetricsFull({
-      read: 0,
-      created: 0,
-      total: 0,
-      hitRate: null,
-      supported: false,
-    })
-    expect(out).toBe('[Cache: N/A]')
-  })
-})
-
-describe('hit-rate edge cases (plan-mandated coverage)', () => {
-  test('0 read / 0 created on supported provider → hitRate = 0 (not null) when total > 0', () => {
-    const m = extractCacheMetrics({ input_tokens: 500 }, 'anthropic')
-    expect(m.read).toBe(0)
-    expect(m.created).toBe(0)
-    expect(m.hitRate).toBe(0)
-  })
-
-  test('read only (no created) computes proportion correctly', () => {
-    const m = extractCacheMetrics(
-      { input_tokens: 0, cache_read_input_tokens: 800, cache_creation_input_tokens: 0 },
-      'anthropic',
-    )
-    expect(m.read).toBe(800)
-    expect(m.created).toBe(0)
-    expect(m.total).toBe(800)
-    expect(m.hitRate).toBe(1)
-  })
-
-  test('created only (first turn — no reads yet) gives 0 hit rate', () => {
-    const m = extractCacheMetrics(
-      {
-        input_tokens: 200,
-        cache_read_input_tokens: 0,
-        cache_creation_input_tokens: 1_000,
-      },
-      'anthropic',
-    )
-    expect(m.read).toBe(0)
-    expect(m.created).toBe(1_000)
-    expect(m.total).toBe(1_200)
-    expect(m.hitRate).toBe(0)
-  })
-
-  test('mixed read + created + fresh input — full denominator', () => {
-    const m = extractCacheMetrics(
-      {
-        input_tokens: 500,
-        cache_read_input_tokens: 3_000,
-        cache_creation_input_tokens: 1_500,
-      },
-      'anthropic',
-    )
-    // Denominator = fresh(500) + created(1500) + read(3000) = 5_000
-    // Hit = read/total = 3000 / 5000 = 0.6
-    expect(m.total).toBe(5_000)
-    expect(m.hitRate).toBe(0.6)
-  })
-
-  test('N/A (unsupported provider) preserves null hit-rate even with populated usage', () => {
-    // Simulate a Copilot usage payload that might look like OpenAI shape —
-    // we must NOT try to read it and must report supported:false.
-    const m = extractCacheMetrics(
-      { prompt_tokens: 5_000, prompt_tokens_details: { cached_tokens: 2_000 } },
-      'copilot',
-    )
-    expect(m.supported).toBe(false)
-    expect(m.read).toBe(0)
-    expect(m.hitRate).toBeNull()
-  })
-})
-
-describe('addCacheMetrics — session aggregation', () => {
-  test('sums read/created/total and recomputes hit rate', () => {
-    const a = {
-      read: 100,
-      created: 50,
-      total: 300,
-      hitRate: 100 / 300,
-      supported: true,
-    }
-    const b = {
-      read: 200,
-      created: 0,
-      total: 400,
-      hitRate: 0.5,
-      supported: true,
-    }
-    const sum = addCacheMetrics(a, b)
-    expect(sum.read).toBe(300)
-    expect(sum.created).toBe(50)
-    expect(sum.total).toBe(700)
-    expect(sum.hitRate).toBeCloseTo(300 / 700, 5)
-  })
-
-  test('unsupported + supported = supported (so we never lose honest data)', () => {
-    const unsupported = {
-      read: 0,
-      created: 0,
-      total: 0,
-      hitRate: null,
-      supported: false,
-    }
-    const supported = {
-      read: 10,
-      created: 0,
-      total: 100,
-      hitRate: 0.1,
-      supported: true,
-    }
-    expect(addCacheMetrics(unsupported, supported)).toBe(supported)
-    expect(addCacheMetrics(supported, unsupported)).toBe(supported)
-  })
-
-  test('unsupported + unsupported = unsupported', () => {
-    const u = {
-      read: 0,
-      created: 0,
-      total: 0,
-      hitRate: null,
-      supported: false,
-    }
-    const sum = addCacheMetrics(u, u)
-    expect(sum.supported).toBe(false)
-  })
-})